<main>
<div class=”blog-content-box”>
<div class=”article-header-box”>
<div class=”article-header”>
<div class=”article-title-box”>
<span class=”article-type type-2 float-left”>转</span> <h1 class=”title-article”>Hadoop性能调优、YARN的内存和CPU配置</h1>
</div>
<div class=”article-info-box”>
<div class=”article-bar-top”>
<span class=”time”>2018年06月12日 21:01:54</span>
<a class=”follow-nickName” href=”https://me.csdn.net/toto1297488504″ target=”_blank”>toto1297488504</a>
<span class=”read-count”>阅读数：2417</span>
</div>
<div class=”operating”>
</div>
</div>
</div>
</div>
<article class=”baidu_pl”>
<div id=”article_content” class=”article_content clearfix csdn-tracking-statistics” data-pid=”blog” data-mod=”popu_307″ data-dsm=”post”>
<div id=”content_views” class=”markdown_views”>
<!– flowchart 箭头图标 勿删 –>
<svg xmlns=”http://www.w3.org/2000/svg” style=”display: none;”><path stroke-linecap=”round” d=”M5,0 0,2.5 5,5z” id=”raphael-marker-block” style=”-webkit-tap-highlight-color: rgba(0, 0, 0, 0);”></path></svg>
<p>转： <br>
<a href=”https://blog.csdn.net/dehu_zhou/article/details/52808752″ rel=”nofollow” target=”_blank”>https://blog.csdn.net/dehu_zhou/article/details/52808752</a> <br>
<a href=”https://blog.csdn.net/dxl342/article/details/52840455″ rel=”nofollow” target=”_blank”>https://blog.csdn.net/dxl342/article/details/52840455</a></p>
<p>Hadoop为用户作业提供了多种可配置的参数，以允许用户根据作业特点调整这些参数值使作业运行效率达到最优。</p>
<h1 id=”一-应用程序编写规范”><a name=”t0″></a>一 应用程序编写规范</h1>
<h2 id=”1设置combiner”><a name=”t1″></a>1.设置Combiner</h2>
<p>对于一大批MapReduce程序，如果可以设置一个Combiner,那么对于提高作业性能是十分有帮助的。Combiner可减少Map Task中间输出的结果，从而减少各个Reduce Task的远程拷贝数据量，最终表现为Map Task和Reduce Task执行时间缩短。</p>
<h2 id=”2-选择合理的writable类型”><a name=”t2″></a>2. 选择合理的Writable类型</h2>
<p>在MapReduce模型中，Map Task和Reduce Task的输入和输出类型均为Writable。Hadoop本身已经提供了很多Writable实现，包括IntWritable、FloatWritable。为应用程序处理的数据选择合适的Writable类型可大大提升性能。比如处理整数类型数据时，直接采用IntWritable比先以Text类型读入在转换为整数类型要高效。如果输出整数的大部分可用一个或两个字节保存，那么直接采用VIntWritable或者VLongWritable，它们采用了变长整型的编码方式，可以大大减少输出数据量。</p>
<h1 id=”二-作业级别参数调优”><a name=”t3″></a>二 作业级别参数调优</h1>
<h2 id=”1规划合理的任务数目”><a name=”t4″></a>1.规划合理的任务数目</h2>
<p>在Hadoop中，每个Map Task处理一个Input Split。Input Split的划分方式是由用户自定义的InputFormat决定的，默认情况下，有以下三个参数决定。 <br>
mapred.min.split.size ：Input Split的最小值 默认值1 <br>
mapred.max.split.szie: Input Split的最大值 <br>
dfs.block.size：HDFS 中一个block大小 默认值64MB <br>
golsize:它是用户期望的Input Split数目=totalSize/numSplits ,其中totalSize为文件的总大小；numSplits为用户设定的Map Task个数，默认情况下是1. <br>
splitSize = max{minSize,min{goalSize,blockSize}} 如果想让InputSize尺寸大于block尺寸，直接增大配置参数mpared.min.split.size即可。</p>
<h2 id=”2增加输入文件的副本数”><a name=”t5″></a>2.增加输入文件的副本数</h2>
<p>如果一个作业并行执行的任务数目非常多，那么这些任务共同的输入文件可能成为瓶颈。为防止多个任务并行读取一个文件内容造成瓶颈，用户可根据需要增加输入文件的副本数目。</p>
<h2 id=”3启动推测执行机制”><a name=”t6″></a>3.启动推测执行机制</h2>
<p>推测执行是Hadoop对“拖后腿”的任务的一种优化机制，当一个作业的某些任务运行速度明显慢于同作业的其他任务时，Hadoop会在另一个节点上为“慢任务”启动一个备份任务，这样两个任务同时处理一份数据，而Hadoop最终会将优先完成的那个任务的结果作为最终结果，并将另一个任务杀掉。</p>
<h2 id=”4设置失败容忍度”><a name=”t7″></a>4.设置失败容忍度</h2>
<p>Hadoop运行设置任务级别和作业级别的失败容忍度。作业级别的失败容忍度是指Hadoop允许每个作业有一定比例的任务运行失败，这部分任务对应的输入数据将被忽略； <br>
任务级别的失败容忍度是指Hadoop允许任务失败后再在另外节点上尝试运行，如果一个任务经过若干次尝试运行后仍然运行失败，那么Hadoop才会最终认为该任务运行失败。 <br>
用户应该根据应用程序的特点设置合理的失败容忍度，以尽快让作业运行完成和避免没必要的资源浪费。</p>
<h2 id=”5适当打开jvm重用功能”><a name=”t8″></a>5.适当打开JVM重用功能</h2>
<p>为了实现任务隔离，Hadoop将每个任务放到一个单独的JVM中执行，而对于执行时间较短的任务，JVM启动和关闭的时间将占用很大比例时间，为此，用户可以启用JVM重用功能，这样一个JVM可连续启动多个同类型的任务。</p>
<h2 id=”6设置任务超时时间”><a name=”t9″></a>6.设置任务超时时间</h2>
<p>如果一个任务在一定的时间内未汇报进度，则TaskTracker会主动将其杀死，从而在另一个节点上重新启动执行。用户可根据实际需要配置任务超时时间。</p>
<h2 id=”7合理使用distributedcache”><a name=”t10″></a>7.合理使用DistributedCache</h2>
<p>一般情况下，得到外部文件有两种方法：一种是外部文件与应用程序jar包一起放到客户端，当提交作业时由客户端上传到HDFS的一个目录下，然后通过Distributed Cache分发到各个节点上；另一种方法是事先将外部文件直接放到HDFS上，从效率上讲，第二种方法更高效。第二种方法不仅节省了客户端上传文件的时间，还隐含着告诉DistributedCache:”请将文件下载到各个节点的pubic级别共享目录中”，这样，后续所有的作业可重用已经下载好的文件，不必重复下载。</p>
<h2 id=”8跳过坏记录”><a name=”t11″></a>8.跳过坏记录</h2>
<p>Hadoop为用户提供了跳过坏记录的功能，当一条或几条坏数据记录导致任务运行失败时，Hadoop可自动识别并跳过这些坏记录。</p>
<h2 id=”9提高作业优先级”><a name=”t12″></a>9.提高作业优先级</h2>
<p>所有Hadoop作业调度器进行任务调度时均会考虑作业优先级这一因素。作业的优先级越高，它能够获取的资源（slot数目)也越多。Hadoop提供了5种作业优先级，分别为VERY_HIGH、 HIGH、 NORMAL、 LOW、 VERY_LOW。 <br>
注：在生产环境中，管理员已经按照作业重要程度对作业进行了分级，不同重要程度的作业允许配置的优先级不同，用户可以擅自进行调整。</p>
<h2 id=”10合理控制reduce-task的启动时机”><a name=”t13″></a>10.合理控制Reduce Task的启动时机</h2>
<p>如果Reduce Task启动过早，则可能由于Reduce Task长时间占用Reduce slot资源造成”slot Hoarding”现象，从而降低资源利用率；反之，如果Reduce Task启动过晚，则会导致Reduce Task获取资源延迟，增加了作业的运行时间。</p>
<h1 id=”三-任务级别参数调优”><a name=”t14″></a>三 任务级别参数调优</h1>
<p>hadoop任务级别参数调优分两个方面: Map Task和Reduce Task。</p>
<h2 id=”1map-task调优”><a name=”t15″></a>1.Map Task调优</h2>
<pre class=”prettyprint”><code class=”has-numbering”>map运行阶段分为:Read、Map、Collect、Spill、Merge五个阶段。
map 任务执行会产生中间数据,但这些中间结果并没有直接IO到磁盘上,而是先存储在缓存(buffer)中,并在缓存中进行一些预排序来优化整个map的性能,存储map中间数据的缓存默认大小为100M，由io.sort.mb 参数指定。这个大小可以根据需要调整。当map任务产生了非常大的中间数据时可以适当调大该参数,使缓存能容纳更多的map中间数据，而不至于大频率的IO磁盘,当系统性能的瓶颈在磁盘IO的速度上,可以适当的调大此参数来减少频繁的IO带来的性能障碍。
由于map任务运行时中间结果首先存储在缓存中,默认当缓存的使用量达到80%(或0.8)的时候就开始写入磁盘,这个过程叫做spill(也叫溢出),进行spill的缓存大小可以通过io.sort.spill.percent 参数调整，这个参数可以影响spill的频率。进而可以影响IO的频率。
当map任务计算成功完成之后，如果map任务有输出，则会产生多个spill。接下来map必须将些spill进行合并,这个过程叫做merge, merge过程是并行处理spill的,每次并行多少个spill是由参数io.sort.factor指定的默认为10个。但是当spill的数量非常大的时候，merge一次并行运行的spill仍然为10个,这样仍然会频繁的IO处理,因此适当的调大每次并行处理的spill数有利于减少merge数因此可以影响map的性能。
当map输出中间结果的时候也可以配置压缩。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li></ul></pre>
<p><img src=”https://img-blog.csdn.net/20180612204931390?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””> </p>
<h2 id=”2-reduce-task调优”><a name=”t16″></a>2. Reduce Task调优</h2>
<pre class=”prettyprint”><code class=”has-numbering”>reduce 运行阶段分为shuflle(copy) merge sort reduce write五个阶段。
shuffle 阶段为reduce 全面拷贝map任务成功结束之后产生的中间结果,如果上面map任务采用了压缩的方式,那么reduce 将map任务中间结果拷贝过来后首先进行解压缩,这一切是在reduce的缓存中做的,当然也会占用一部分cpu。为了优化reduce的执行时间,reduce也不是等到所有的map数据都拷贝过来的时候才开始运行reduce任务，而是当job执行完第一个map任务时开始运行的。reduce 在shuffle阶段 实际上是从不同的并且已经完成的map上去下载属于自己的数据,由于map任务数很多,所有这个copy过程是并行的,既同时有许多个reduce取拷贝map，这个并行的线程是通过mapred.reduce.parallel.copies 参数指定，默认为5个,也就是说无论map的任务数是多少个，默认情况下一次只能有5个reduce的线程去拷贝map任务的执行结果。所以当map任务数很多的情况下可以适当的调整该参数，这样可以让reduce快速的获得运行数据来完成任务。
reduce线程在下载map数据的时候也可能因为各种各样的原因(网络原因、系统原因等），存储该map数据所在的datannode 发生了故障，这种情况下reduce任务将得不到该datanode上的数据了,同时该 download thread 会尝试从别的datanode下载,可以通过mapred.reduce.copy.backoff (默认为30秒)来调整下载线程的下载时间，如果网络不好的集群可以通过增加该参数的值来增加下载时间,以免因为下载时间过长reduce将该线程判断为下载失败。
reduce 下载线程在map结果下载到本地时,由于是多线程并行下载，所以也需要对下载回来的数据进行merge,所以map阶段设置的io.sort.factor 也同样会影响这个reduce的。
同map一样 该缓冲区大小也不是等到完全被占满的时候才写入磁盘而是默认当完成0.66的时候就开始写磁盘操作,该参数是通过mapred.job.shuffle.merge.percent 指定的。
当reduce 开始进行计算的时候通过mapred.job.reduce.input.buffer.percent 来指定需要多少的内存百分比来作为reduce读已经sort好的数据的buffer百分比,该值默认为0。Hadoop假设用户的reduce()函数需要所有的JVM内存，因此执行reduce()函数前要释放所有内存。如果设置了该值，可将部分文件保存在内存中(不必写到磁盘上)。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li><li style=”color: rgb(153, 153, 153);”>6</li></ul></pre>
<p><img src=”https://img-blog.csdn.net/20180612205603918?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””></p>
<pre class=”prettyprint”><code class=”has-numbering”>总之，Map Task和Reduce Task调优的一个原则就是减少数据的传输量、尽量使用内存、减少磁盘IO的次数、增大任务并行数，除此之外还有根据自己集群及网络的实际情况来调优。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li></ul></pre>
<h1 id=”三-管理员角度调优”><a name=”t17″></a>三 管理员角度调优</h1>
<pre class=”prettyprint”><code class=”has-numbering”>管理员负责为用户作业提供一个高效的运行环境。管理员需要从全局出发，通过调整一些关键参数提高系统的吞吐率和性能。总体上来看，管理员需从硬件选择、操作系统参数调优、JVM参数调优和Hadoop参数调优等四个角度入手，为Hadoop用户提供一个高效的作业运行环境。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li></ul></pre>
<p>硬件选择 <br>
Hadoop自身架构的基本特点决定了其硬件配置的选项。Hadoop采用了Master/Slave架构，其中，master维护了全局元数据信息，重要性远远大于slave。在较低Hadoop版本中，master存在单点故障问题，因此，master的配置应远远好于各个slave。</p>
<p><strong>操作系统参数调优</strong></p>
<h2 id=”1增大同时打开的文件描述符和网络连接上限”><a name=”t18″></a>1.增大同时打开的文件描述符和网络连接上限</h2>
<pre class=”prettyprint”><code class=”has-numbering”>使用ulimit命令将允许同时打开的文件描述符数目上限增大至一个合适的值。同时调整内核参数net.core.somaxconn网络连接数目至一个足够大的值。
补充：net.core.somaxconn的作用
net.core.somaxconn是Linux中的一个kernel参数，表示socket监听（listen）的backlog上限。什么是backlog呢？backlog就是socket的监听队列，当一个请求（request）尚未被处理或建立时，它会进入backlog。而socket server可以一次性处理backlog中的所有请求，处理后的请求不再位于监听队列中。当server处理请求较慢，以至于监听队列被填满后，新来的请求会被拒绝。在Hadoop 1.0中，参数ipc.server.listen.queue.size控制了服务端socket的监听队列长度，即backlog长度，默认值是128。而Linux的参数net.core.somaxconn默认值同样为128。当服务端繁忙时，如NameNode或JobTracker，128是远远不够的。这样就需要增大backlog，例如我们的3000台集群就将ipc.server.listen.queue.size设成了32768，为了使得整个参数达到预期效果，同样需要将kernel参数net.core.somaxconn设成一个大于等于32768的值。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li></ul></pre>
<h2 id=”2关闭swap分区”><a name=”t19″></a>2.关闭swap分区</h2>
<pre class=”prettyprint”><code class=”has-numbering”>避免使用swap分区，提供程序的执行效率。
除此之外，设置合理的预读取缓冲区的大小、文件系统选择与配置及I/O调度器选择等
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li></ul></pre>
<p><strong>JVM参数调优</strong> <br>
由于Hadoop中的每个服务和任务均会运行在一个单独的JVM中，因此，JVM的一些重要参数也会影响Hadoop性能。管理员可通过调整JVM FLAGS和JVM垃圾回收机制提高Hadoop性能。</p>
<p><strong>Hadoop参数调优</strong></p>
<h3 id=”1合理规划资源”><a name=”t20″></a>1.合理规划资源</h3>
<pre class=”prettyprint”><code class=”has-numbering”>设置合理的槽位数目
在Hadoop中，计算资源是用槽位表示的。slot分为两种：Map Slot和Reduce Slot。每种slot代表一定量的资源，且同种slot是同质的，也就是说，同种slot代表的资源量是相同的。管理员需要根据实际需要为TaskTracker配置一定数目的Map Slot和Reduce Slot数目，从而限制每个TaskTracker上并发执行的Map Task和Reduce Task的数目。
编写健康监测脚本
Hadoop允许管理员为每个TaskTracker配置一个节点健康状况监测脚本。TaskTracker中包含一个专门的线程周期性执行该脚本，并将脚本执行结果通过心跳机制汇报给JobTracker。一旦JobTracker发现某个TaskTracker的当前状况为“不健康”，则会将其加入黑名单，从此不再为它分配任务。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li></ul></pre>
<h3 id=”2-调整心跳配置”><a name=”t21″></a>2. 调整心跳配置</h3>
<pre class=”prettyprint”><code class=”has-numbering”>调整心跳的间隔 因根据自己集群的规模适度的调整心跳间隔
启用带外心跳 为了减少任务分配延迟，Hadoop引入了带外心跳。带外心跳不同于常规心跳，它是任务运行结束或者任务运行失败时触发的，能够在出现空闲资源时第一时间通知JobTracker,以便它能够迅速为空闲资源分配新的任务。
除此之外，还包括磁盘块配置、设置合理的RPC Handler和HTTP线程数目、慎用黑名单机制、启用批量任务调度、选择合适的压缩算法、启用预读取机制等。
注：当一个集群的规模较小时，如果一定数量的节点被频繁的加入系统黑名单中，则会大大降低集群的吞吐率和计算能力。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li></ul></pre>
<h1 id=”四yarn的内存和cpu配置”><a name=”t22″></a>四：YARN的内存和CPU配置</h1>
<p>Hadoop YARN同时支持内存和CPU两种资源的调度，本文介绍如何配置YARN对内存和CPU的使用。</p>
<p>YARN作为一个资源调度器，应该考虑到集群里面每一台机子的计算资源，然后根据application申请的资源进行分配Container。Container是YARN里面资源分配的基本单位，具有一定的内存以及CPU资源。</p>
<p>在YARN集群中，平衡内存、CPU、磁盘的资源的很重要的，根据经验，每两个container使用一块磁盘以及一个CPU核的时候可以使集群的资源得到一个比较好的利用。</p>
<h2 id=”1内存配置”><a name=”t23″></a>1、内存配置</h2>
<p>关于 内存 相关的配置可以参考hortonwork公司的文档 Determine HDP Memory Configuration Settings （<a href=”https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.1.1/bk_installing_manually_book/content/rpm-chap1-11.html” rel=”nofollow” target=”_blank”>https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.1.1/bk_installing_manually_book/content/rpm-chap1-11.html</a>）来配置你的集群。</p>
<p>YARN以及MAPREDUCE所有可用的内存资源应该要除去系统运行需要的以及其他的hadoop的一些程序，总共保留的内存=系统内存+HBASE内存。</p>
<p>可以参考下面的表格确定应该保留的内存： <br>
<img src=”https://img-blog.csdn.net/20180613155007579?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””></p>
<p>计算每台机子最多可以拥有多少个container，可以使用下面的公式: <br>
containers = min (2*CORES, 1.8*DISKS, (Total available RAM) / MIN_CONTAINER_SIZE) <br>
说明：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs mathematica has-numbering”>CORES 为机器CPU核数
DISKS 为机器上挂载的磁盘个数
<span class=”hljs-keyword”>Total</span> available RAM 为机器总内存
MIN_CONTAINER_SIZE 是指container最小的容量大小，这需要根据具体情况去设置，可以参考下面的表格</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li></ul></pre>
<p><img src=”https://img-blog.csdn.net/20180613155229427?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””></p>
<p>每个container的平均使用内存大小计算方式为：</p>
<p>RAM-per-container = max(MIN_CONTAINER_SIZE, (Total Available RAM) / containers))</p>
<p>通过上面的计算，YARN以及MAPREDUCE可以这样配置： <br>
<img src=”https://img-blog.csdn.net/20180613155324928?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””></p>
<p>举个例子：对于128G内存、32核CPU的机器，挂载了7个磁盘，根据上面的说明，系统保留内存为24G，不适应HBase情况下，系统剩余可用内存为104G，计算containers值如下：</p>
<p>containers = min (2*32, 1.8* 7 , (128-24)/2) = min (64, 12.6 , 51) = 13</p>
<p>计算RAM-per-container值如下：</p>
<p>RAM-per-container = max (2, (124-24)/13) = max (2, 8) = 8</p>
<p>这样集群中下面的参数配置值如下： <br>
<img src=”https://img-blog.csdn.net/20180613155420773?watermark/2/text/aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3RvdG8xMjk3NDg4NTA0/font/5a6L5L2T/fontsize/400/fill/I0JBQkFCMA==/dissolve/70″ alt=”这里写图片描述” title=””></p>
<p>你也可以使用脚本 yarn-utils.py （<a href=”https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.1.1/bk_installing_manually_book/content/rpm-chap1-9.html” rel=”nofollow” target=”_blank”>https://docs.hortonworks.com/HDPDocuments/HDP2/HDP-2.1.1/bk_installing_manually_book/content/rpm-chap1-9.html</a>）来计算上面的值：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs lasso has-numbering”>python yarn<span class=”hljs-attribute”>-utils</span><span class=”hljs-built_in”>.</span>py <span class=”hljs-attribute”>-c</span> <span class=”hljs-number”>32</span> <span class=”hljs-attribute”>-m</span> <span class=”hljs-number”>128</span> <span class=”hljs-attribute”>-d</span> <span class=”hljs-number”>7</span> <span class=”hljs-attribute”>-k</span> <span class=”hljs-literal”>False</span></code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li></ul></pre>
<p>返回结果如下：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs avrasm has-numbering”>Using cores=<span class=”hljs-number”>32</span> memory=<span class=”hljs-number”>128</span>GB disks=<span class=”hljs-number”>7</span> hbase=False
Profile: cores=<span class=”hljs-number”>32</span> memory=<span class=”hljs-number”>106496</span>MB reserved=<span class=”hljs-number”>24</span>GB usableMem=<span class=”hljs-number”>104</span>GB disks=<span class=”hljs-number”>7</span>
Num Container=<span class=”hljs-number”>13</span>
Container Ram=<span class=”hljs-number”>8192</span>MB
Used Ram=<span class=”hljs-number”>104</span>GB
Unused Ram=<span class=”hljs-number”>24</span>GB
yarn<span class=”hljs-preprocessor”>.scheduler</span><span class=”hljs-preprocessor”>.minimum</span>-allocation-mb=<span class=”hljs-number”>8192</span>
yarn<span class=”hljs-preprocessor”>.scheduler</span><span class=”hljs-preprocessor”>.maximum</span>-allocation-mb=<span class=”hljs-number”>106496</span>
yarn<span class=”hljs-preprocessor”>.nodemanager</span><span class=”hljs-preprocessor”>.resource</span><span class=”hljs-preprocessor”>.memory</span>-mb=<span class=”hljs-number”>106496</span>
mapreduce<span class=”hljs-preprocessor”>.map</span><span class=”hljs-preprocessor”>.memory</span><span class=”hljs-preprocessor”>.mb</span>=<span class=”hljs-number”>8192</span>
mapreduce<span class=”hljs-preprocessor”>.map</span><span class=”hljs-preprocessor”>.java</span><span class=”hljs-preprocessor”>.opts</span>=-Xmx6553m
mapreduce<span class=”hljs-preprocessor”>.reduce</span><span class=”hljs-preprocessor”>.memory</span><span class=”hljs-preprocessor”>.mb</span>=<span class=”hljs-number”>8192</span>
mapreduce<span class=”hljs-preprocessor”>.reduce</span><span class=”hljs-preprocessor”>.java</span><span class=”hljs-preprocessor”>.opts</span>=-Xmx6553m
yarn<span class=”hljs-preprocessor”>.app</span><span class=”hljs-preprocessor”>.mapreduce</span><span class=”hljs-preprocessor”>.am</span><span class=”hljs-preprocessor”>.resource</span><span class=”hljs-preprocessor”>.mb</span>=<span class=”hljs-number”>8192</span>
yarn<span class=”hljs-preprocessor”>.app</span><span class=”hljs-preprocessor”>.mapreduce</span><span class=”hljs-preprocessor”>.am</span><span class=”hljs-preprocessor”>.command</span>-opts=-Xmx6553m
mapreduce<span class=”hljs-preprocessor”>.task</span><span class=”hljs-preprocessor”>.io</span><span class=”hljs-preprocessor”>.sort</span><span class=”hljs-preprocessor”>.mb</span>=<span class=”hljs-number”>3276</span></code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li><li style=”color: rgb(153, 153, 153);”>6</li><li style=”color: rgb(153, 153, 153);”>7</li><li style=”color: rgb(153, 153, 153);”>8</li><li style=”color: rgb(153, 153, 153);”>9</li><li style=”color: rgb(153, 153, 153);”>10</li><li style=”color: rgb(153, 153, 153);”>11</li><li style=”color: rgb(153, 153, 153);”>12</li><li style=”color: rgb(153, 153, 153);”>13</li><li style=”color: rgb(153, 153, 153);”>14</li><li style=”color: rgb(153, 153, 153);”>15</li><li style=”color: rgb(153, 153, 153);”>16</li></ul></pre>
<p>对应的xml配置为：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs xml has-numbering”><span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.nodemanager.resource.memory-mb<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>106496<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.scheduler.minimum-allocation-mb<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>8192<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.scheduler.maximum-allocation-mb<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>106496<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.app.mapreduce.am.resource.mb<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>8192<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.app.mapreduce.am.command-opts<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>-Xmx6553m<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span></code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li><li style=”color: rgb(153, 153, 153);”>6</li><li style=”color: rgb(153, 153, 153);”>7</li><li style=”color: rgb(153, 153, 153);”>8</li><li style=”color: rgb(153, 153, 153);”>9</li><li style=”color: rgb(153, 153, 153);”>10</li><li style=”color: rgb(153, 153, 153);”>11</li><li style=”color: rgb(153, 153, 153);”>12</li><li style=”color: rgb(153, 153, 153);”>13</li><li style=”color: rgb(153, 153, 153);”>14</li><li style=”color: rgb(153, 153, 153);”>15</li><li style=”color: rgb(153, 153, 153);”>16</li><li style=”color: rgb(153, 153, 153);”>17</li><li style=”color: rgb(153, 153, 153);”>18</li><li style=”color: rgb(153, 153, 153);”>19</li><li style=”color: rgb(153, 153, 153);”>20</li></ul></pre>
<p>另外，还有一下几个参数：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs lasso has-numbering”>yarn<span class=”hljs-built_in”>.</span>nodemanager<span class=”hljs-built_in”>.</span>vmem<span class=”hljs-attribute”>-pmem</span><span class=”hljs-attribute”>-ratio</span> ：任务每使用<span class=”hljs-number”>1</span>MB物理内存，最多可使用虚拟内存量，默认是<span class=”hljs-number”>2.1</span>。
yarn<span class=”hljs-built_in”>.</span>nodemanager<span class=”hljs-built_in”>.</span>pmem<span class=”hljs-attribute”>-check</span><span class=”hljs-attribute”>-enabled</span> ：是否启动一个线程检查每个任务正使用的物理内存量，如果任务超出分配值，则直接将其杀掉，默认是<span class=”hljs-literal”>true</span>。
yarn<span class=”hljs-built_in”>.</span>nodemanager<span class=”hljs-built_in”>.</span>vmem<span class=”hljs-attribute”>-pmem</span><span class=”hljs-attribute”>-ratio</span> ：是否启动一个线程检查每个任务正使用的虚拟内存量，如果任务超出分配值，则直接将其杀掉，默认是<span class=”hljs-literal”>true</span>。</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li></ul></pre>
<p>第一个参数的意思是当一个map任务总共分配的物理内存为8G的时候，该任务的container最多内分配的堆内存为6.4G，可以分配的虚拟内存上限为8*2.1=16.8G。另外，照这样算下去，每个节点上YARN可以启动的Map数为104/8=13个，似乎偏少了，这主要是和我们挂载的磁盘数太少了有关，人为的调整 RAM-per-container 的值为4G或者更小的一个值是否更合理呢？当然，这个要监控集群实际运行情况来决定了。 <br>
<strong>CPU配置</strong> <br>
YARN中目前的CPU被划分成虚拟CPU（CPU virtual Core），这里的虚拟CPU是YARN自己引入的概念，初衷是，考虑到不同节点的CPU性能可能不同，每个CPU具有的计算能力也是不一样的，比如某个物理CPU的计算能力可能是另外一个物理CPU的2倍，这时候，你可以通过为第一个物理CPU多配置几个虚拟CPU弥补这种差异。用户提交作业时，可以指定每个任务需要的虚拟CPU个数。</p>
<p>在YARN中，CPU相关配置参数如下：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs avrasm has-numbering”>yarn<span class=”hljs-preprocessor”>.nodemanager</span><span class=”hljs-preprocessor”>.resource</span><span class=”hljs-preprocessor”>.cpu</span>-vcores ：表示该节点上YARN可使用的虚拟CPU个数，默认是<span class=”hljs-number”>8</span>，注意，目前推荐将该值设值为与物理CPU核数数目相同。如果你的节点CPU核数不够<span class=”hljs-number”>8</span>个，则需要调减小这个值，而YARN不会智能的探测节点的物理CPU总数。
yarn<span class=”hljs-preprocessor”>.scheduler</span><span class=”hljs-preprocessor”>.minimum</span>-allocation-vcores ：单个任务可申请的最小虚拟CPU个数，默认是<span class=”hljs-number”>1</span>，如果一个任务申请的CPU个数少于该数，则该对应的值改为这个数。
yarn<span class=”hljs-preprocessor”>.scheduler</span><span class=”hljs-preprocessor”>.maximum</span>-allocation-vcores ：单个任务可申请的最多虚拟CPU个数，默认是<span class=”hljs-number”>32</span>。</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li></ul></pre>
<p>对于一个CPU核数较多的集群来说，上面的默认配置显然是不合适的，在我的测试集群中，4个节点每个机器CPU核数为32，可以配置为：</p>
<pre class=”prettyprint” name=”code”><code class=”hljs xml has-numbering”><span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.nodemanager.resource.cpu-vcores<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>32<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>property</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>name</span>></span>yarn.scheduler.maximum-allocation-vcores<span class=”hljs-tag”></<span class=”hljs-title”>name</span>></span>
<span class=”hljs-tag”><<span class=”hljs-title”>value</span>></span>128<span class=”hljs-tag”></<span class=”hljs-title”>value</span>></span>
<span class=”hljs-tag”></<span class=”hljs-title”>property</span>></span></code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li><li style=”color: rgb(153, 153, 153);”>3</li><li style=”color: rgb(153, 153, 153);”>4</li><li style=”color: rgb(153, 153, 153);”>5</li><li style=”color: rgb(153, 153, 153);”>6</li><li style=”color: rgb(153, 153, 153);”>7</li><li style=”color: rgb(153, 153, 153);”>8</li></ul></pre>
<p><strong>总结</strong> <br>
根据上面的说明，我的测试集群中集群节点指标如下： <br>
每个节点分配的物理内存、虚拟内存和CPU核数如下： <br>
实际生产环境中，可能不会像上面那样设置，比如不会把所有节点的CPU核数都分配给Spark，需要保留一个核留给系统使用；另外，内存上限也会做些设置。</p>
<h1 id=”小结”><a name=”t24″></a>小结</h1>
<pre class=”prettyprint”><code class=”has-numbering”>Hadoop 性能调优是一项工程浩大的工作，它不仅涉及Hadoop本身的性能调优，还涉及更底层的硬件、操作系统和Java虚拟机等系统的调优。
总体来说，提高作业运行效率需要Hadoop管理员和作业拥有者共同的努力，其中，管理员负责为用户提供一个高效的作业运行环境，而用户则根据自己作业的特点让它尽可能快地运行完成。
</code><ul class=”pre-numbering” style=””><li style=”color: rgb(153, 153, 153);”>1</li><li style=”color: rgb(153, 153, 153);”>2</li></ul></pre> </div>
<link href=”https://csdnimg.cn/release/phoenix/mdeditor/markdown_views-7b4cdcb592.css” rel=”stylesheet”>
</div>
</article>
</div>
<script>
(function(){
function collection(){
if (currentUserName) {
if (!$(this).hasClass(“liked”)) {
$.ajax({
url: ‘https://my.csdn.net/my/favorite/do_add/2’,
dataType: ‘json’,
type: ‘POST’,
xhrFields: {
withCredentials: true
},
data: {
title: articleTit,
url: curentUrl,
share: 1,
map_name: ”
},
success: function(data) {
if (data.succ == 1) {
$(‘.btn-bookmark’).addClass(“liked”);
$(‘.article-footer-bookmark-btn’).addClass(“liked”).children(‘span’).text(‘已收藏’);
alert(‘收藏成功,可以在个人中心查看我的收藏’);
} else {
if (data.msg === “您已经收藏过”) {
$(‘.btn-bookmark’).addClass(“liked”);
$(‘.article-footer-bookmark-btn’).addClass(“liked”).children(‘span’).text(‘已收藏’);
}
alert(data.msg);
}
}
});
} else {
alert(‘您已经收藏过’);
}
} else {
window.csdn.loginBox.show();
}
}
window.csdn = window.csdn ? window.csdn : {};
window.csdn.articleCollection = collection;
function setArticleH(btnReadmore,posi){
var winH = $(window).height();
var articleBox = $(“div.article_content”);
var artH = articleBox.height();
if(artH > winH*posi){
articleBox.css({
‘height’:winH*posi+’px’,
‘overflow’:’hidden’
})
btnReadmore.click(function(){
articleBox.removeAttr(“style”);
$(this).parent().remove()
})
}else{
btnReadmore.parent().remove()
}
}
var btnReadmore = $(“#btn-readmore”);
$(‘.article-footer-bookmark-btn’).click(window.csdn.articleCollection)
if(btnReadmore.length>0){
if(currentUserName){
setArticleH(btnReadmore,3);
}else{
setArticleH(btnReadmore,1.2);
}
}else{
$(‘.hide-article-box’).addClass(‘hide-article-style’);
}
})()
</script>
<script>
$(“.MathJax”).remove();
if($(‘div.markdown_views pre.prettyprint code.hljs’).length > 0 ){
$(‘div.markdown_views’)[0].className = ‘markdown_views’;
}
</script>
<a id=”commentBox”></a>
<div class=”comment-box”>
<div class=”comment-edit-box d-flex”>
<a id=”commentsedit”></a>
<div class=”user-img”>
<a href=”//me.csdn.net/qq_41641810″ target=”_blank”>
<img class=”” src=”https://avatar.csdn.net/C/5/F/3_qq_41641810.jpg”>
</a>
</div>
<form id=”commentform”>
<input type=”hidden” id=”comment_replyId”>
<textarea class=”comment-content” name=”comment_content” id=”comment_content” placeholder=”想对作者说点什么”></textarea>
<div class=”opt-box”> <!– d-flex –>
<div id=”ubbtools” class=”add_code”>
<a href=”#insertcode” code=”code” target=”_self”><i class=”icon iconfont icon-daima”></i></a>
</div>
<input type=”hidden” id=”comment_replyId” name=”comment_replyId”>
<input type=”hidden” id=”article_id” name=”article_id” value=”80671128″>
<input type=”hidden” id=”comment_userId” name=”comment_userId” value=””>
<input type=”hidden” id=”commentId” name=”commentId” value=””>
<div style=”display: none;” class=”csdn-tracking-statistics tracking-click” data-mod=”popu_384″><a href=”#” target=”_blank” class=”comment_area_btn”>发表评论</a></div>
<div class=”dropdown” id=”myDrap”>
<a class=”dropdown-face d-flex align-items-center” data-toggle=”dropdown” role=”button” aria-haspopup=”true” aria-expanded=”false”>
<div class=”txt-selected text-truncate”>添加代码片</div>
<svg class=”icon d-block” aria-hidden=”true”>
<use xmlns:xlink=”http://www.w3.org/1999/xlink” xlink:href=”#csdnc-triangledown”></use>
</svg>
</a>
<ul class=”dropdown-menu” id=”commentCode” aria-labelledby=”drop4″>
<li><a data-code=”html”>HTML/XML</a></li>
<li><a data-code=”objc”>objective-c</a></li>
<li><a data-code=”ruby”>Ruby</a></li>
<li><a data-code=”php”>PHP</a></li>
<li><a data-code=”csharp”>C</a></li>
<li><a data-code=”cpp”>C++</a></li>
<li><a data-code=”javascript”>JavaScript</a></li>
<li><a data-code=”python”>Python</a></li>
<li><a data-code=”java”>Java</a></li>
<li><a data-code=”css”>CSS</a></li>
<li><a data-code=”sql”>SQL</a></li>
<li><a data-code=”plain”>其它</a></li>
</ul>
</div>
<div class=”right-box”>
<span id=”tip_comment” class=”tip”>还能输入<em>1000</em>个字符</span>
<input type=”submit” class=”btn btn-sm btn-red btn-comment” value=”发表评论”>
</div>
</div>
</form>
</div>
<div class=”comment-list-container”>
<a id=”comments”></a>
<div class=”comment-list-box” style=”max-height: none;”><ul class=”comment-list”><li class=”comment-line-box d-flex” data-commentid=”9047340″ data-replyname=”weixin_44120720″> <a target=”_blank” href=”https://me.csdn.net/weixin_44120720″><img src=”https://avatar.csdn.net/0/5/8/3_weixin_44120720.jpg” alt=”weixin_44120720″ class=”avatar”></a> <div class=”right-box “> <div class=”info-box”> <a target=”_blank” href=”https://me.csdn.net/weixin_44120720″><span class=”name “>淩辰0c0：</span></a> <span class=”comment”>楼主大大，CDH版的配置也是这样吗</span><span class=”date” title=”2019-01-15 00:34:58″>(1个月前</span><span class=”floor-num”>#2楼)</span><span class=”opt-box”><a class=”btn btn-link-blue btn-report” data-type=”report”>举报</a><a class=”btn btn-link-blue btn-reply” data-type=”reply”>回复</a></span></div> </div> </li></ul><ul class=”comment-list”><li class=”comment-line-box d-flex” data-commentid=”9047272″ data-replyname=”weixin_44120720″> <a target=”_blank” href=”https://me.csdn.net/weixin_44120720″><img src=”https://avatar.csdn.net/0/5/8/3_weixin_44120720.jpg” alt=”weixin_44120720″ class=”avatar”></a> <div class=”right-box “> <div class=”info-box”> <a target=”_blank” href=”https://me.csdn.net/weixin_44120720″><span class=”name “>淩辰0c0：</span></a> <span class=”comment”>您好，现在的现在版本的Hadoop 内存配置还是这种机制吗？这篇文档16年的啊</span><span class=”date” title=”2019-01-15 00:22:42″>(1个月前</span><span class=”floor-num”>#1楼)</span><span class=”opt-box”><a class=”btn btn-link-blue btn-report” data-type=”report”>举报</a><a class=”btn btn-link-blue btn-reply” data-type=”reply”>回复</a></span></div> </div> </li></ul></div>
<div id=”commentPage” class=”pagination-box d-none” style=”display: block;”><div id=”Paging_08203955896173589″ class=”ui-paging-container”><ul><li class=”js-page-first js-page-action ui-pager ui-pager-disabled”></li><li class=”js-page-prev js-page-action ui-pager ui-pager-disabled”>上一页</li><li data-page=”1″ class=”ui-pager focus”>1</li><li class=”js-page-next js-page-action ui-pager ui-pager-disabled”>下一页</li><li class=”js-page-last js-page-action ui-pager ui-pager-disabled”></li></ul></div></div>
</div>
</div>
<div class=”recommend-box”>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/lykangjia/article/details/52839497,BlogCommendFromBaidu_1,index_0"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/lykangjia/article/details/52839497″ target=”_blank” title=”Hadoop性能调优总结”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em><em>性能调优</em>总结 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-17</span>
<span class=”read-num hover-hide”>
阅读数
2674</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/lykangjia/article/details/52839497″ target=”_blank” title=”Hadoop性能调优总结”>
<span class=”desc oneline”>目的随着企业要处理的数据量越来越大，Hadoop运行在越来越多的集群上，同时MapReduce由于具有高可扩展性和容错性，已经逐步广泛使用开来。因此也产生很多问题，尤其是性能方面的问题。这里从管理员角…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/lykangjia”>来自： <span class=”blog_title”> lykangjia的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/hy245120020/article/details/75202058,BlogCommendFromBaidu_3,index_1"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/hy245120020/article/details/75202058″ target=”_blank” title=”hadoop性能调优与运维”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em><em>性能调优</em>与运维 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>07-16</span>
<span class=”read-num hover-hide”>
阅读数
3535</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/hy245120020/article/details/75202058″ target=”_blank” title=”hadoop性能调优与运维”>
<span class=”desc oneline”>hadoop性能调优与运维硬件选择操作系统调优与JVM调优hadoop参数调优hive性能调优hadoop运维硬件选择hadoop运行环境如何选择合适的硬件主从节点可靠性：主节点可靠性要好于从节点单节…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/hy245120020″>来自： <span class=”blog_title”> 那年花下月如雪</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/lingbo229/article/details/79054370,BlogCommendFromBaidu_4,index_2"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/lingbo229/article/details/79054370″ target=”_blank” title=”YARN的Memory和CPU调优配置详解”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>YARN</em>的Memory和<em>CPU</em>调优<em>配置</em>详解 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-13</span>
<span class=”read-num hover-hide”>
阅读数
1052</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/lingbo229/article/details/79054370″ target=”_blank” title=”YARN的Memory和CPU调优配置详解”>
<span class=”desc oneline”>HadoopYARN同时支持内存和CPU两种资源的调度，本文介绍如何配置YARN对内存和CPU的使用。YARN作为一个资源调度器，应该考虑到集群里面每一台机子的计算资源，然后根据application…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/lingbo229″>来自： <span class=”blog_title”> 微步的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/gx181710/article/details/70187080,BlogCommendFromBaidu_0,index_3"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/gx181710/article/details/70187080″ target=”_blank” title=”Hadoop 性能优化”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em> 性能优化 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>04-15</span>
<span class=”read-num hover-hide”>
阅读数
5549</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/gx181710/article/details/70187080″ target=”_blank” title=”Hadoop 性能优化”>
<span class=”desc oneline”>Hive性能优化1.概述 继续《那些年使用Hive踩过的坑》一文中的剩余部分，本篇博客赘述了在工作中总结Hive的常用优化手段和在工作中使用Hive出现的问题。下面开始本篇文章的优化介绍。2.介绍 …</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/gx181710″>来自： <span class=”blog_title”> AMY行致远的IT</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_36932624/article/details/81503702,BlogCommendFromBaidu_2,index_4"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36932624/article/details/81503702″ target=”_blank” title=”Hadoop性能调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em><em>性能调优</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-08</span>
<span class=”read-num hover-hide”>
阅读数
66</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36932624/article/details/81503702″ target=”_blank” title=”Hadoop性能调优”>
<span class=”desc oneline”>Hadoop为用户作业提供了多种可配置的参数，以允许用户根据作业特点调整这些参数值使作业运行效率达到最优。 一应用程序编写规范1.设置Combiner 对于一大批MapReduce程序，如果可以…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_36932624″>来自： <span class=”blog_title”> 邢小成</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_19917081/article/details/54575644,BlogCommendFromBaidu_5,index_5"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_19917081/article/details/54575644″ target=”_blank” title=”hdfs+yarn 参数调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
hdfs+<em>yarn</em> 参数调优 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-16</span>
<span class=”read-num hover-hide”>
阅读数
5201</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_19917081/article/details/54575644″ target=”_blank” title=”hdfs+yarn 参数调优”>
<span class=”desc oneline”>1.系统参数调整中文名称英文名称默认值推荐值或推荐策略修改命令作用备注socket的监听队列最大值net.core.somaxconn1281024或更大echo1024>/proc/sys/net/…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_19917081″>来自： <span class=”blog_title”> qq_19917081的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_blog clearfix” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/slq1023/article/details/49407043,BlogCommendFromBaidu_6,index_6"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/slq1023/article/details/49407043″ target=”_blank” title=”第118讲：Hadoop内存需求量及Namenode内存使用详解学习笔记”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
第118讲：<em>Hadoop</em><em>内存</em>需求量及Namenode<em>内存</em>使用详解学习笔记 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-25</span>
<span class=”read-num hover-hide”>
阅读数
4359</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/slq1023/article/details/49407043″ target=”_blank” title=”第118讲：Hadoop内存需求量及Namenode内存使用详解学习笔记”>
<span class=”desc oneline”>第118讲：Hadoop内存需求量及Namenode内存使用详解学习笔记本期内容：1.hadoop的内存使用2.namenode内存使用详解namenode内存大小涉及到集群的规模。hadoop默认为…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/slq1023″>来自： <span class=”blog_title”> slq1023的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/panhongan/article/details/42474627,BlogCommendFromBaidu_7,index_7"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/panhongan/article/details/42474627,BlogCommendFromBaidu_7,index_7"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/panhongan/article/details/42474627″ target=”_blank” title=”hadoop yarn的内存和cpu参数配置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em> <em>yarn</em>的<em>内存</em>和<em>cpu</em>参数<em>配置</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-06</span>
<span class=”read-num hover-hide”>
阅读数
650</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/panhongan/article/details/42474627″ target=”_blank” title=”hadoop yarn的内存和cpu参数配置”>
<span class=”desc oneline”>【导读】还在整理中</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/panhongan”>来自： <span class=”blog_title”> panhongan的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/GreatElite/article/details/25559859,BlogCommendFromBaidu_8,index_8"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/GreatElite/article/details/25559859,BlogCommendFromBaidu_8,index_8"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/GreatElite/article/details/25559859″ target=”_blank” title=”Hadoop集群硬件标准配置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>集群硬件标准<em>配置</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>05-11</span>
<span class=”read-num hover-hide”>
阅读数
2482</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/GreatElite/article/details/25559859″ target=”_blank” title=”Hadoop集群硬件标准配置”>
<span class=”desc oneline”>在我们选择硬件的时候，往往需要考虑应用得到的性能和经费开支。为此，我们必须在满足实际需要和经济可行上找到一个完美的平衡。下面，以Hadoop集群应用为了例子，说明需要考虑的要素。 1.I/O绑定的工…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/GreatElite”>来自： <span class=”blog_title”> 用心做事</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/q383965374/article/details/79429495,searchFromBaidu_9,index_9"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/q383965374/article/details/79429495,searchFromBaidu_9,index_9"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/q383965374/article/details/79429495″ target=”_blank” title=”hadoop进阶—hadoop性能优化(一)—hdfs空间不足的管理优化”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em>进阶—<em>hadoop</em>性能优化(一)—hdfs空间不足的管理优化 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-03</span>
<span class=”read-num hover-hide”>
阅读数
1635</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/q383965374/article/details/79429495″ target=”_blank” title=”hadoop进阶—hadoop性能优化(一)—hdfs空间不足的管理优化”>
<span class=”desc oneline”>Hadoop空间不足，hive首先就会没法跑了，进度始终是0%。将HDFS备份数降低将默认的备份数3设置为2。步骤：CDH–&amp;gt;HDFS–&amp;gt;配置–&amp;gt;搜索dfs….</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/q383965374″>来自： <span class=”blog_title”> 直到世界的尽头</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box blog-expert-recommend-box” style=”display: block;”>
<div class=”d-flex”>
<div class=”blog-expert-recommend”>
<div class=”blog-expert”>
<div class=”blog-expert-flexbox” data-track-view=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><div class=”blog-expert-item”><div class=”blog-expert-info-box”><div class=”blog-expert-img-box” data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/duan_zhihua” target=”_blank”><img src=”https://avatar.csdn.net/3/8/F/3_duan_zhihua.jpg” alt=”段智华” title=”段智华”><svg class=”icon” aria-hidden=”true”><use xmlns:xlink=”http://www.w3.org/1999/xlink” xlink:href=”#csdnc-blogexpert”></use></svg></a><span data-track-click=”{"mod":"popu_710","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><span class=”blog-expert-button-follow btn-red-follow” data-name=”duan_zhihua” data-nick=”段智华”>关注</span></span></div><div class=”info”><span data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/duan_zhihua” target=”_blank”><h5 class=”oneline” title=”段智华”>段智华</h5></a></span> <p></p><p class=”article-num” title=”864篇文章”> 864篇文章</p><p class=”article-num” title=”排名:1000+”> 排名:1000+</p><p></p></div></div></div><div class=”blog-expert-item”><div class=”blog-expert-info-box”><div class=”blog-expert-img-box” data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/Androidlushangderen” target=”_blank”><img src=”https://avatar.csdn.net/9/F/0/3_androidlushangderen.jpg” alt=”Android路上的人” title=”Android路上的人”><svg class=”icon” aria-hidden=”true”><use xmlns:xlink=”http://www.w3.org/1999/xlink” xlink:href=”#csdnc-blogexpert”></use></svg></a><span data-track-click=”{"mod":"popu_710","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><span class=”blog-expert-button-follow btn-red-follow” data-name=”Androidlushangderen” data-nick=”Android路上的人”>关注</span></span></div><div class=”info”><span data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/Androidlushangderen” target=”_blank”><h5 class=”oneline” title=”Android路上的人”>Android路上的人</h5></a></span> <p></p><p class=”article-num” title=”316篇文章”> 316篇文章</p><p class=”article-num” title=”排名:1000+”> 排名:1000+</p><p></p></div></div></div><div class=”blog-expert-item”><div class=”blog-expert-info-box”><div class=”blog-expert-img-box” data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/ZYC88888″ target=”_blank”><img src=”https://avatar.csdn.net/C/3/C/3_zyc88888.jpg” alt=”ZhaoYingChao88″ title=”ZhaoYingChao88″><svg class=”icon” aria-hidden=”true”><use xmlns:xlink=”http://www.w3.org/1999/xlink” xlink:href=”#csdnc-blogexpert”></use></svg></a><span data-track-click=”{"mod":"popu_710","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><span class=”blog-expert-button-follow btn-red-follow” data-name=”ZYC88888″ data-nick=”ZhaoYingChao88″>关注</span></span></div><div class=”info”><span data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/ZYC88888″ target=”_blank”><h5 class=”oneline” title=”ZhaoYingChao88″>ZhaoYingChao88</h5></a></span> <p></p><p class=”article-num” title=”395篇文章”> 395篇文章</p><p class=”article-num” title=”排名:6000+”> 排名:6000+</p><p></p></div></div></div><div class=”blog-expert-item”><div class=”blog-expert-info-box”><div class=”blog-expert-img-box” data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/pany8125″ target=”_blank”><img src=”https://avatar.csdn.net/C/1/5/3_pany8125.jpg” alt=”pany8125″ title=”pany8125″></a><span data-track-click=”{"mod":"popu_710","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><span class=”blog-expert-button-follow btn-red-follow” data-name=”pany8125″ data-nick=”pany8125″>关注</span></span></div><div class=”info”><span data-track-click=”{"mod":"popu_709","con":"https://blog.csdn.net/tototuzuoquan/article/details/80671128"}”><a href=”https://blog.csdn.net/pany8125″ target=”_blank”><h5 class=”oneline” title=”pany8125″>pany8125</h5></a></span> <p></p><p class=”article-num” title=”42篇文章”> 42篇文章</p><p class=”article-num” title=”排名:千里之外”> 排名:千里之外</p><p></p></div></div></div></div>
</div>
</div>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_26033611/article/details/79891282,searchFromBaidu_10,index_10"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_26033611/article/details/79891282,searchFromBaidu_10,index_10"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_26033611/article/details/79891282″ target=”_blank” title=”【工程类】【Hadoop】性能调优方法”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
【工程类】【<em>Hadoop</em>】<em>性能调优</em>方法 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>04-11</span>
<span class=”read-num hover-hide”>
阅读数
71</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_26033611/article/details/79891282″ target=”_blank” title=”【工程类】【Hadoop】性能调优方法”>
<span class=”desc oneline”>当我们写完一个MR程序之后，我们希望能够加快程序的执行速度，那么怎么进行性能的调优呢，有如下几种方法？1、mapper的数量适当调整mapper的数量，使得每个mapper的运行时间在1分钟为宜。因为…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_26033611″>来自： <span class=”blog_title”> 空空如也</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u010010664/article/details/77066031,BlogCommendFromGuangxin_11,index_11"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u010010664/article/details/77066031,BlogCommendFromGuangxin_11,index_11"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u010010664/article/details/77066031″ target=”_blank” title=”Hive on Spark调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Hive on Spark调优 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-10</span>
<span class=”read-num hover-hide”>
阅读数
2114</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u010010664/article/details/77066031″ target=”_blank” title=”Hive on Spark调优”>
<span class=”desc oneline”>之前在HiveonSpark跑TPCx-BB测试时，100g的数据量要跑十几个小时，一看CPU和内存的监控，发现 POWER_TEST阶段（依次执行30个查询）CPU只用了百分之十几，也就是没有把整个…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u010010664″>来自： <span class=”blog_title”> 窗外的屋檐</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/weixin_38750084/article/details/82948121,BlogCommendFromGuangxin_12,index_12"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/weixin_38750084/article/details/82948121,BlogCommendFromGuangxin_12,index_12"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/weixin_38750084/article/details/82948121″ target=”_blank” title=”hadoop：搭建Hadoop集群，一个月6T的数量需要几台服务器”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em>：搭建<em>Hadoop</em>集群，一个月6T的数量需要几台服务器 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-05</span>
<span class=”read-num hover-hide”>
阅读数
377</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/weixin_38750084/article/details/82948121″ target=”_blank” title=”hadoop：搭建Hadoop集群，一个月6T的数量需要几台服务器”>
<span class=”desc oneline”>配置几个namenode,几个datanode,namenode和datanode怎么部署，内存与硬盘大小？？ 最好是两个做成HA 关于硬盘: 6T的数据容量,看你副本数量设置是多少,一…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/weixin_38750084″>来自： <span class=”blog_title”> 不花的花和尚的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_37142346/article/details/81490314,BlogCommendFromGuangxin_13,index_13"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_37142346/article/details/81490314,BlogCommendFromGuangxin_13,index_13"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_37142346/article/details/81490314″ target=”_blank” title=”Spark性能调优之合理分配系统资源以及并行度的调节”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Spark<em>性能调优</em>之合理分配系统资源以及并行度的调节 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-07</span>
<span class=”read-num hover-hide”>
阅读数
269</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_37142346/article/details/81490314″ target=”_blank” title=”Spark性能调优之合理分配系统资源以及并行度的调节”>
<span class=”desc oneline”>在Spark系统中，调优是非常重要。首先，调优的第一步也是首先要考虑的就是系统的资源，只有在合理、有效的分配完了系统的资源，才会去考虑其他调优点从而提升系统的性能与速度。虽然对于系统的资源的分配，看上…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_37142346″>来自： <span class=”blog_title”> 不清不慎的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/DZRYWYBL/article/details/79413077,BlogCommendFromGuangxin_14,index_14"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/DZRYWYBL/article/details/79413077,BlogCommendFromGuangxin_14,index_14"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/DZRYWYBL/article/details/79413077″ target=”_blank” title=”Ubuntu14.4安装QQ”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Ubuntu14.4安装QQ </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-01</span>
<span class=”read-num hover-hide”>
阅读数
996</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/DZRYWYBL/article/details/79413077″ target=”_blank” title=”Ubuntu14.4安装QQ”>
<span class=”desc oneline”>Ctrl+Alt+t打开终端： 1、Ubuntu软件中心搜索wine并安装。 命令安装：sudoapt-getinstallwine 2、下载qq：WineQQ7.8-20151109-L…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/DZRYWYBL”>来自： <span class=”blog_title”> Somnus陳的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/pany8125/article/details/47955175,BlogCommendFromQuerySearch_15,index_15"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/pany8125/article/details/47955175,BlogCommendFromQuerySearch_15,index_15"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/pany8125/article/details/47955175″ target=”_blank” title=”CDH集群调优：内存、Vcores和DRF”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
CDH集群调优：<em>内存</em>、Vcores和DRF </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-24</span>
<span class=”read-num hover-hide”>
阅读数
3339</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/pany8125/article/details/47955175″ target=”_blank” title=”CDH集群调优：内存、Vcores和DRF”>
<span class=”desc oneline”>原文URL：http://blog.selfup.cn/1631.html#comment-403吐槽最近“闲”来无事，通过CM把vcores使用情况调出来看了一眼，发现不论集群中有多少个任务在跑，已…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/pany8125″>来自： <span class=”blog_title”> 前进的南山</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/zyj8170/article/details/48491903,BlogCommendFromQuerySearch_16,index_16"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/zyj8170/article/details/48491903,BlogCommendFromQuerySearch_16,index_16"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zyj8170/article/details/48491903″ target=”_blank” title=”Hadoop YARN如何调度内存和CPU”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em> <em>YARN</em>如何调度<em>内存</em>和<em>CPU</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-16</span>
<span class=”read-num hover-hide”>
阅读数
704</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zyj8170/article/details/48491903″ target=”_blank” title=”Hadoop YARN如何调度内存和CPU”>
<span class=”desc oneline”>HadoopYARN同时支持内存和CPU两种资源的调度（默认只支持内存，如果想进一步调度CPU，需要自己进行一些配置），本文将介绍YARN是如何对这些资源进行调度和隔离的。在YARN中，资源管理由Re…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/zyj8170″>来自： <span class=”blog_title”> zyj8170的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/luofazha2012/article/details/80287031,BlogCommendFromQuerySearch_17,index_17"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/luofazha2012/article/details/80287031,BlogCommendFromQuerySearch_17,index_17"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/luofazha2012/article/details/80287031″ target=”_blank” title=”[Hadoop性能调优]”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
[<em>Hadoop</em><em>性能调优</em>] </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>05-11</span>
<span class=”read-num hover-hide”>
阅读数
9141</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/luofazha2012/article/details/80287031″ target=”_blank” title=”[Hadoop性能调优]”>
<span class=”desc oneline”>一、Hadoop概述 随着企业要处理的数据量越来越大，MapReduce思想越来越受到重视。Hadoop是MapReduce的一个开源实现，由于其良好的扩展性和容错性，已得到越来越广泛的应用。 Ha…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/luofazha2012″>来自： <span class=”blog_title”> 程序车轮</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/duan_zhihua/article/details/72884491,BlogCommendFromQuerySearch_18,index_18"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/duan_zhihua/article/details/72884491,BlogCommendFromQuerySearch_18,index_18"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/duan_zhihua/article/details/72884491″ target=”_blank” title=”第1章 对运行在YARN上的Spark进行性能调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
第1章 对运行在<em>YARN</em>上的Spark进行<em>性能调优</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-06</span>
<span class=”read-num hover-hide”>
阅读数
1312</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/duan_zhihua/article/details/72884491″ target=”_blank” title=”第1章 对运行在YARN上的Spark进行性能调优”>
<span class=”desc oneline”>第1章 对运行在YARN上的Spark进行性能调优1.1 运行环境Jar包管理及数据本地性原理调优实践1.1.1运行环境Jar包管理及和数据本地性原理在YARN上运行Spark需要在Spar…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/duan_zhihua”>来自： <span class=”blog_title”> 段智华的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/love_lixr/article/details/81454548,BlogCommendFromQuerySearch_19,index_19"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/love_lixr/article/details/81454548,BlogCommendFromQuerySearch_19,index_19"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/love_lixr/article/details/81454548″ target=”_blank” title=”yarn进程CPU占用率百分之百”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>yarn</em>进程<em>CPU</em>占用率百分之百 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-06</span>
<span class=”read-num hover-hide”>
阅读数
670</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/love_lixr/article/details/81454548″ target=”_blank” title=”yarn进程CPU占用率百分之百”>
<span class=”desc oneline”>阿里云出现挖矿程序Linux根绝Pid查看进程名top命令查看占用CPU高的进程ps-aux|grepPID号ps-ef|grepPID号查看yarn程序的信息ps-ef|grepyarnyarn46…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/love_lixr”>来自： <span class=”blog_title”> love_lixr的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/pansaky/article/details/83347357,BlogCommendFromBaidu_20,index_20"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/pansaky/article/details/83347357,BlogCommendFromBaidu_20,index_20"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/pansaky/article/details/83347357″ target=”_blank” title=”Hadoop性能调优全面总结”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em><em>性能调优</em>全面总结 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-24</span>
<span class=”read-num hover-hide”>
阅读数
583</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/pansaky/article/details/83347357″ target=”_blank” title=”Hadoop性能调优全面总结”>
<span class=”desc oneline”>一、Hadoop概述随着企业要处理的数据量越来越大，MapReduce思想越来越受到重视。Hadoop是MapReduce的一个开源实现，由于其良好的扩展性和容错性，已得到越来越广泛的应用。Hadoo…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/pansaky”>来自： <span class=”blog_title”> pansaky的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/dajuezhao/article/details/6591034,BlogCommendFromBaidu_21,index_21"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/dajuezhao/article/details/6591034,BlogCommendFromBaidu_21,index_21"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/dajuezhao/article/details/6591034″ target=”_blank” title=”Hadoop集群的配置调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>集群的<em>配置</em>调优 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>07-07</span>
<span class=”read-num hover-hide”>
阅读数
3</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/dajuezhao/article/details/6591034″ target=”_blank” title=”Hadoop集群的配置调优”>
<span class=”desc oneline”>一、背景Hadoop的集群使用也有些时候了，不过都是小集群的使用（数量小于30台）。在这个过程中不断的进行着配置调优的操作。最早的时候，在网上也没有找到一些合适的配置调优方法，于是，我在这里列举一些基…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/dajuezhao”>来自： <span class=”blog_title”> dajuezhao的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/javastart/article/details/44175293,BlogCommendFromBaidu_22,index_22"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/javastart/article/details/44175293,BlogCommendFromBaidu_22,index_22"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/javastart/article/details/44175293″ target=”_blank” title=”为Hadoop集群选择合适的硬件配置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
为<em>Hadoop</em>集群选择合适的硬件<em>配置</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-10</span>
<span class=”read-num hover-hide”>
阅读数
6740</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/javastart/article/details/44175293″ target=”_blank” title=”为Hadoop集群选择合适的硬件配置”>
<span class=”desc oneline”>随着ApacheHadoop的起步，云客户的增多面临的首要问题就是如何为他们新的的Hadoop集群选择合适的硬件。尽管Hadoop被设计为运行在行业标准的硬件上，提出一个理想的集群配置不想提供硬件规格…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/javastart”>来自： <span class=”blog_title”> 张伟的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident recommend-download-box clearfix” data-track-view=”{"mod":"popu_614","con":",https://download.csdn.net/download/xjdlg/9861973,BlogCommendFromBaidu_23,index_23"}” data-track-click=”{"mod":"popu_614","con":",https://download.csdn.net/download/xjdlg/9861973,BlogCommendFromBaidu_23,index_23"}” data-flg=”true”>
<a href=”https://download.csdn.net/download/xjdlg/9861973″ target=”_blank”>
<div class=”content clearfix”>
<div class=””>
<h4 class=”text-truncate oneline clearfix”>
<em>hadoop</em> <em>yarn</em>优化<em>配置</em>项生成工具 </h4>
<span class=”data float-right”>06-06</span>
</div>
<div class=”desc oneline”>
hadoop yarn优化配置项生成工具，需要安装python环境，执行例如：python yarn-utils.py -c 24 -m 256 -d 4 </div>
<span class=”type-show type-show-download”>下载</span>
</div>
</a>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/aaa1117a8w5s6d/article/details/43192885,BlogCommendFromBaidu_24,index_24"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/aaa1117a8w5s6d/article/details/43192885,BlogCommendFromBaidu_24,index_24"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/aaa1117a8w5s6d/article/details/43192885″ target=”_blank” title=”查看yarn的container的资源占用情况”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
查看<em>yarn</em>的container的资源占用情况 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-27</span>
<span class=”read-num hover-hide”>
阅读数
7050</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/aaa1117a8w5s6d/article/details/43192885″ target=”_blank” title=”查看yarn的container的资源占用情况”>
<span class=”desc oneline”>yarn上运行程序是由container来执行的，那么当我们想要知道每台节点对应的container的使用情况时，需要从何入手呢？开始我以为yarn的系统命令会有相应的提示，于是yarn–help，…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/aaa1117a8w5s6d”>来自： <span class=”blog_title”> aaa1117a8w5s6d的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u010226454/article/details/80907088,BlogCommendFromBaidu_25,index_25"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u010226454/article/details/80907088,BlogCommendFromBaidu_25,index_25"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u010226454/article/details/80907088″ target=”_blank” title=”Yarn Container计算和MapReduce的内存配置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Yarn</em> Container计算和MapReduce的<em>内存</em><em>配置</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>07-04</span>
<span class=”read-num hover-hide”>
阅读数
430</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u010226454/article/details/80907088″ target=”_blank” title=”Yarn Container计算和MapReduce的内存配置”>
<span class=”desc oneline”>在Hadoop2.0中,YARN负责管理MapReduce中的资源(内存,CPU等)并且将其打包成Container.这样可以精简MapReduce,使之专注于其擅长的数据处理任务,将无需考虑资源调度…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u010226454″>来自： <span class=”blog_title”> 王茗灏的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/zhao897426182/article/details/78353079,BlogCommendFromQuerySearch_26,index_26"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/zhao897426182/article/details/78353079,BlogCommendFromQuerySearch_26,index_26"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhao897426182/article/details/78353079″ target=”_blank” title=”Yarn的Memory和CPU调优配置详解”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Yarn</em>的Memory和<em>CPU</em>调优<em>配置</em>详解 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-26</span>
<span class=”read-num hover-hide”>
阅读数
396</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhao897426182/article/details/78353079″ target=”_blank” title=”Yarn的Memory和CPU调优配置详解”>
<span class=”desc oneline”>PPHadoopYarn同事支持内存和CPU两种资源的调度，本文介绍如何配置YARN对内存和CPU的使用。Yarn作为一个资源调度器，应该考虑到集群里面每一台机子的计算资源，然后根据applicati…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/zhao897426182″>来自： <span class=”blog_title”> 就问你吃不吃药</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/zhanglh046/article/details/78567070,BlogCommendFromQuerySearch_27,index_27"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/zhanglh046/article/details/78567070,BlogCommendFromQuerySearch_27,index_27"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhanglh046/article/details/78567070″ target=”_blank” title=”MapReduce之如何给运行在YARN上的MapReduce作业配置内存”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
MapReduce之如何给运行在<em>YARN</em>上的MapReduce作业<em>配置</em><em>内存</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-18</span>
<span class=”read-num hover-hide”>
阅读数
1169</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhanglh046/article/details/78567070″ target=”_blank” title=”MapReduce之如何给运行在YARN上的MapReduce作业配置内存”>
<span class=”desc oneline”>关于mapreduce程序运行在yarn上时内存的分配一直是一个让我蒙圈的事情，单独查任何一个资料都不能很好的理解透彻。于是，最近查了大量的资料，综合各种解释，终于理解到了一个比较清晰的程度，在这里将…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/zhanglh046″>来自： <span class=”blog_title”> happy19870612’s blog</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/xiaolong_4_2/article/details/84849220,BlogCommendFromQuerySearch_28,index_28"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/xiaolong_4_2/article/details/84849220,BlogCommendFromQuerySearch_28,index_28"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xiaolong_4_2/article/details/84849220″ target=”_blank” title=”CDH集群yarn内存调优”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
CDH集群<em>yarn</em><em>内存</em>调优 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>12-06</span>
<span class=”read-num hover-hide”>
阅读数
208</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xiaolong_4_2/article/details/84849220″ target=”_blank” title=”CDH集群yarn内存调优”>
<span class=”desc oneline”>原文地址：http://blog.selfup.cn/1631.html?utm_source=tuicool&amp;amp;utm_medium=referral 吐槽 最近“闲”来无事，通过CM…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/xiaolong_4_2″>来自： <span class=”blog_title”> aboth的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_25302531/article/details/80623791,BlogCommendFromQuerySearch_29,index_29"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_25302531/article/details/80623791,BlogCommendFromQuerySearch_29,index_29"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_25302531/article/details/80623791″ target=”_blank” title=”浅谈YARN中Container容器（内存、CPU分配）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
浅谈<em>YARN</em>中Container容器（<em>内存</em>、<em>CPU</em>分配） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-08</span>
<span class=”read-num hover-hide”>
阅读数
2731</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_25302531/article/details/80623791″ target=”_blank” title=”浅谈YARN中Container容器（内存、CPU分配）”>
<span class=”desc oneline”> 前言：hadoop2.x版本和hadoop1.x版本的一个区别就是：hadoop1.x中负责资源和作业调度的是MapReduce，hadoop2.x版本后，MapReduce只专注于计算，资源…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_25302531″>来自： <span class=”blog_title”> 青青青的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/hthou/article/details/53195319,BlogCommendFromQuerySearch_30,index_30"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/hthou/article/details/53195319,BlogCommendFromQuerySearch_30,index_30"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/hthou/article/details/53195319″ target=”_blank” title=”spark+yarn调整计算任务的内存占用”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
spark+<em>yarn</em>调整计算任务的<em>内存</em>占用 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-17</span>
<span class=”read-num hover-hide”>
阅读数
1966</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/hthou/article/details/53195319″ target=”_blank” title=”spark+yarn调整计算任务的内存占用”>
<span class=”desc oneline”>调整的背景公司正式环境有一个小小的spark+hadoop集群，yarn的可用内存只有小小的40G，五个节点，每个节点8G内存，因此需要对每个上线的计算任务占用的资源都要精打细算。写了两个sparks…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/hthou”>来自： <span class=”blog_title”> 无聊小刚的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/chndata/article/details/42234443,BlogCommendFromBaidu_31,index_31"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/chndata/article/details/42234443,BlogCommendFromBaidu_31,index_31"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/chndata/article/details/42234443″ target=”_blank” title=”Hadoop性能调优学习概述”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em><em>性能调优</em>学习概述 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>12-29</span>
<span class=”read-num hover-hide”>
阅读数
2391</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/chndata/article/details/42234443″ target=”_blank” title=”Hadoop性能调优学习概述”>
<span class=”desc oneline”>目的随着企业要处理的数据量越来越大，Hadoop运行在越来越多的集群上，同时MapReduce由于具有高可扩展性和容错性，已经逐步广泛使用开来。因此也产生很多问题，尤其是性能方面的问题。这里从管理员角…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/chndata”>来自： <span class=”blog_title”> chndata的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/fp196391196391/article/details/8302121,BlogCommendFromBaidu_32,index_32"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/fp196391196391/article/details/8302121,BlogCommendFromBaidu_32,index_32"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/fp196391196391/article/details/8302121″ target=”_blank” title=”hadoop 性能调优 重要参数设置技巧”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em> <em>性能调优</em> 重要参数设置技巧 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>12-16</span>
<span class=”read-num hover-hide”>
阅读数
9441</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/fp196391196391/article/details/8302121″ target=”_blank” title=”hadoop 性能调优 重要参数设置技巧”>
<span class=”desc oneline”>这里主要针对Mapreduce的性能调优。这一两个月在做mapreduce的性能调优，有些心得，还是要记下来的，以郷后人~这里主要涉及的参数包括：HDFS：dfs.block.sizeMapredur…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/fp196391196391″>来自： <span class=”blog_title”> fp196391196391的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/chndata/article/details/46003399,BlogCommendFromBaidu_33,index_33"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/chndata/article/details/46003399,BlogCommendFromBaidu_33,index_33"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/chndata/article/details/46003399″ target=”_blank” title=”Hadoop性能调优总结（一）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em><em>性能调优</em>总结（一） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>05-26</span>
<span class=”read-num hover-hide”>
阅读数
1</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/chndata/article/details/46003399″ target=”_blank” title=”Hadoop性能调优总结（一）”>
<span class=”desc oneline”>目的随着企业要处理的数据量越来越大，Hadoop运行在越来越多的集群上，同时MapReduce由于具有高可扩展性和容错性，已经逐步广泛使用开来。因此也产生很多问题，尤其是性能方面的问题。这里从管理员角…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/chndata”>来自： <span class=”blog_title”> chndata的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/victory0508/article/details/50595276,BlogCommendFromBaidu_34,index_34"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/victory0508/article/details/50595276,BlogCommendFromBaidu_34,index_34"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/victory0508/article/details/50595276″ target=”_blank” title=”Hadoop参数优化整理”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>参数优化整理 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-27</span>
<span class=”read-num hover-hide”>
阅读数
2842</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/victory0508/article/details/50595276″ target=”_blank” title=”Hadoop参数优化整理”>
<span class=”desc oneline”>系统调整1. io.file.buffer.size用来设置缓存的大小。不论是对硬盘或者是网络操作来讲，较大的缓存都可以提供更高的数据传输，但这也就意味着更大的内存消耗和延迟。这个参数要设置为系统页面…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/victory0508″>来自： <span class=”blog_title”> victory0508的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/javastart/article/details/50606734,BlogCommendFromBaidu_35,index_35"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/javastart/article/details/50606734,BlogCommendFromBaidu_35,index_35"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/javastart/article/details/50606734″ target=”_blank” title=”Hadoop平台架构–存储篇”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>平台架构–存储篇 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-29</span>
<span class=”read-num hover-hide”>
阅读数
6959</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/javastart/article/details/50606734″ target=”_blank” title=”Hadoop平台架构–存储篇”>
<span class=”desc oneline”>Hadoop平台架构–存储篇By whoami 发表于2016-01-25文章目录1. 简介2. 走向分布式3. 存储规划4. HDFS目录规划4.1. linuxos目录规划4.2. linux主…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/javastart”>来自： <span class=”blog_title”> 张伟的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/Androidlushangderen/article/details/50002015,BlogCommendFromBaidu_36,index_36"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/Androidlushangderen/article/details/50002015,BlogCommendFromBaidu_36,index_36"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Androidlushangderen/article/details/50002015″ target=”_blank” title=”Hadoop动态调整Map Task内存资源大小”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>动态调整Map Task<em>内存</em>资源大小 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-23</span>
<span class=”read-num hover-hide”>
阅读数
5947</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Androidlushangderen/article/details/50002015″ target=”_blank” title=”Hadoop动态调整Map Task内存资源大小”>
<span class=”desc oneline”>前言我们都知道,在Hadoop中,一个Job的执行需要转化成1个个的Task去执行,在Task中,有会有2个类型,一个为MapTask,另一个就是ReduceTask.当然,这不是最底层的级别,在Ta…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/Androidlushangderen”>来自： <span class=”blog_title”> 走在前往架构师的路上</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u012042963/article/details/53099638,BlogCommendFromBaidu_37,index_37"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u012042963/article/details/53099638,BlogCommendFromBaidu_37,index_37"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u012042963/article/details/53099638″ target=”_blank” title=”Hadoop运行因为container的内存大小而抛错”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>运行因为container的<em>内存</em>大小而抛错 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-09</span>
<span class=”read-num hover-hide”>
阅读数
4477</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u012042963/article/details/53099638″ target=”_blank” title=”Hadoop运行因为container的内存大小而抛错”>
<span class=”desc oneline”>本文转载自：http://blog.chinaunix.net/uid-25691489-id-5587957.htmlHadoop运行中抛出如下异常：Container[pid=41355,cont…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u012042963″>来自： <span class=”blog_title”> LibsKnowledge的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/youngqj/article/details/47315167,BlogCommendFromBaidu_38,index_38"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/youngqj/article/details/47315167,BlogCommendFromBaidu_38,index_38"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/youngqj/article/details/47315167″ target=”_blank” title=”如何判定并设置YARN 和MapReduce 内存”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
如何判定并设置<em>YARN</em> 和MapReduce <em>内存</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-06</span>
<span class=”read-num hover-hide”>
阅读数
4555</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/youngqj/article/details/47315167″ target=”_blank” title=”如何判定并设置YARN 和MapReduce 内存”>
<span class=”desc oneline”>11.1. ManuallyCalculateYARNandMapReduceMemoryConfigurationSettingsThissectiondescribeshowtomanuallyc…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/youngqj”>来自： <span class=”blog_title”> 树蛙PHPER</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/zhangxiong0301/article/details/84723857,BlogCommendFromQuerySearch_39,index_39"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/zhangxiong0301/article/details/84723857,BlogCommendFromQuerySearch_39,index_39"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhangxiong0301/article/details/84723857″ target=”_blank” title=”（转）hadoop yarn 内存相关配置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
（转）<em>hadoop</em> <em>yarn</em> <em>内存</em>相关<em>配置</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-11</span>
<span class=”read-num hover-hide”>
阅读数
8</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/zhangxiong0301/article/details/84723857″ target=”_blank” title=”（转）hadoop yarn 内存相关配置”>
<span class=”desc oneline”>1.YARN中处理能力的基本单元是什么？2.什么是保留内存？3.4到8GContainer建议多少M？在Hadoop2.0中,YARN负责管理MapReduce中的资源(内存,CPU等)并且将其打包成…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/zhangxiong0301″>来自： <span class=”blog_title”> zhangxiong0301的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_36753550/article/details/83065546,BlogCommendFromQuerySearch_40,index_40"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_36753550/article/details/83065546,BlogCommendFromQuerySearch_40,index_40"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36753550/article/details/83065546″ target=”_blank” title=”hadoop 资源配置与队列设置”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em> 资源<em>配置</em>与队列设置 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>10-16</span>
<span class=”read-num hover-hide”>
阅读数
389</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36753550/article/details/83065546″ target=”_blank” title=”hadoop 资源配置与队列设置”>
<span class=”desc oneline”>yarn集群资源设置资源（1）yarn.nodemanager.resource.memory-mb配置nodemanager节点的可用物理内存，默认是8192（MB），如这台服务器有16G可以考虑分…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_36753550″>来自： <span class=”blog_title”> YI依仔的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/leone911/article/details/77849271,BlogCommendFromQuerySearch_41,index_41"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/leone911/article/details/77849271,BlogCommendFromQuerySearch_41,index_41"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/leone911/article/details/77849271″ target=”_blank” title=”yarn使用cgroup隔离cpu资源”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>yarn</em>使用cgroup隔离<em>cpu</em>资源 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-05</span>
<span class=”read-num hover-hide”>
阅读数
1137</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/leone911/article/details/77849271″ target=”_blank” title=”yarn使用cgroup隔离cpu资源”>
<span class=”desc oneline”>yarn默认只管理内存资源,虽然也可以申请cpu资源,但是在没有cpu资源隔离的情况下效果并不是太好.在集群规模大,任务多时资源竞争的问题尤为严重.还好yarn提供的LinuxContainerExe…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/leone911″>来自： <span class=”blog_title”> leone911的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_25302531/article/details/80608288,BlogCommendFromQuerySearch_42,index_42"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_25302531/article/details/80608288,BlogCommendFromQuerySearch_42,index_42"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_25302531/article/details/80608288″ target=”_blank” title=”YARN and MapReduce的内存优化配置详解”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>YARN</em> and MapReduce的<em>内存</em>优化<em>配置</em>详解 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-07</span>
<span class=”read-num hover-hide”>
阅读数
482</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_25302531/article/details/80608288″ target=”_blank” title=”YARN and MapReduce的内存优化配置详解”>
<span class=”desc oneline”>在Hadoop2.x中,YARN负责管理MapReduce中的资源(内存,CPU等)并且将其打包成Container。使之专注于其擅长的数据处理任务,将无需考虑资源调度.如下图所示 …</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_25302531″>来自： <span class=”blog_title”> 青青青的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u014393917/article/details/52621542,BlogCommendFromQuerySearch_43,index_43"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u014393917/article/details/52621542,BlogCommendFromQuerySearch_43,index_43"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u014393917/article/details/52621542″ target=”_blank” title=”yarn中的cgroup调度”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>yarn</em>中的cgroup调度 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-22</span>
<span class=”read-num hover-hide”>
阅读数
2961</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u014393917/article/details/52621542″ target=”_blank” title=”yarn中的cgroup调度”>
<span class=”desc oneline”>yarn中cgroup的调度</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u014393917″>来自： <span class=”blog_title”> 隔壁老杨的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/yaoxiaochuang/article/details/50959158,BlogCommendFromBaidu_44,index_44"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/yaoxiaochuang/article/details/50959158,BlogCommendFromBaidu_44,index_44"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/yaoxiaochuang/article/details/50959158″ target=”_blank” title=”Hadoop实战：*********MapReduce的性能调优（一）*********”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>Hadoop</em>实战：*********MapReduce的<em>性能调优</em>（一）********* </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-22</span>
<span class=”read-num hover-hide”>
阅读数
2779</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/yaoxiaochuang/article/details/50959158″ target=”_blank” title=”Hadoop实战：*********MapReduce的性能调优（一）*********”>
<span class=”desc oneline”>这里主要涉及的参数包括：HDFS：dfs.block.sizeMapredure：io.sort.mbio.sort.spill.percentmapred.local.dirmapred.map.t…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/yaoxiaochuang”>来自： <span class=”blog_title”> 探索地理之源 分享GIS价值</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/xygl2009/article/details/46858671,BlogCommendFromBaidu_45,index_45"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/xygl2009/article/details/46858671,BlogCommendFromBaidu_45,index_45"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xygl2009/article/details/46858671″ target=”_blank” title=”hadoop内存大小设置问题”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>hadoop</em><em>内存</em>大小设置问题 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>07-13</span>
<span class=”read-num hover-hide”>
阅读数
9051</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xygl2009/article/details/46858671″ target=”_blank” title=”hadoop内存大小设置问题”>
<span class=”desc oneline”>前面博客里面提到，运行一次hadoop的时候出现javaheaperror。字面意思分配堆的时候出现错误，我们知道应用程序的动态内存的分配均在堆里面。这里提示堆错误，那必然是内存不够用了。那么这个na…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/xygl2009″>来自： <span class=”blog_title”> 不积跬步，无以致千里；不积小流，无以成江海</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u011031430/article/details/74892130,BlogCommendFromBaidu_46,index_46"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u011031430/article/details/74892130,BlogCommendFromBaidu_46,index_46"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u011031430/article/details/74892130″ target=”_blank” title=”内存小于16GG搭建Hadoop集群的一些个人观点”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
<em>内存</em>小于16GG搭建<em>Hadoop</em>集群的一些个人观点 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>07-09</span>
<span class=”read-num hover-hide”>
阅读数
3637</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u011031430/article/details/74892130″ target=”_blank” title=”内存小于16GG搭建Hadoop集群的一些个人观点”>
<span class=”desc oneline”>很多同鞋在学习大数据过程中，也会在自己电脑上搭建环境，但是内存不足确实是个让人头疼的问题，废话不多说，个人有如下一些想法，分享给大家：1、使用apache原生的hadoop来搭建集群，除了配置繁琐之外…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u011031430″>来自： <span class=”blog_title”> Gavin博客专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/clamaa/article/details/70046185,BlogCommendFromBaidu_47,index_47"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/clamaa/article/details/70046185,BlogCommendFromBaidu_47,index_47"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/clamaa/article/details/70046185″ target=”_blank” title=”一次hadoop集群机器加内存的运维过程”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
一次<em>hadoop</em>集群机器加<em>内存</em>的运维过程 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>04-11</span>
<span class=”read-num hover-hide”>
阅读数
1507</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/clamaa/article/details/70046185″ target=”_blank” title=”一次hadoop集群机器加内存的运维过程”>
<span class=”desc oneline”>由于前期的集群规划问题，导致当前Hadoop集群中的硬件并没有完全利用起来。当前机器的内存CPU比例为2G：1core，但一般的MapReduce任务（数据量处理比较大，逻辑较复杂）的MR两端都需要将…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/clamaa”>来自： <span class=”blog_title”> clamaa的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/iteye_20817/article/details/82241805,BlogCommendFromBaidu_48,index_48"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/iteye_20817/article/details/82241805,BlogCommendFromBaidu_48,index_48"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/iteye_20817/article/details/82241805″ target=”_blank” title=”设置hadoop各个应用的内存”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
设置<em>hadoop</em>各个应用的<em>内存</em> </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-04</span>
<span class=”read-num hover-hide”>
阅读数
21</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/iteye_20817/article/details/82241805″ target=”_blank” title=”设置hadoop各个应用的内存”>
<span class=”desc oneline”>NameNode、Jobtracker、Datanode、Tasktracker设置 conf/hadoop-env.sh 修改HADOOP_HEAPSIZE值，默认为1000MB，标题中四个应用将…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/iteye_20817″>来自： <span class=”blog_title”> The Big Data Way</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/fansy1990/article/details/54314249,BlogCommendFromBaidu_49,index_49"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/fansy1990/article/details/54314249,BlogCommendFromBaidu_49,index_49"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/fansy1990/article/details/54314249″ target=”_blank” title=”Spark On YARN内存和CPU分配”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Spark On <em>YARN</em><em>内存</em>和<em>CPU</em>分配 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-10</span>
<span class=”read-num hover-hide”>
阅读数
1</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/fansy1990/article/details/54314249″ target=”_blank” title=”Spark On YARN内存和CPU分配”>
<span class=”desc oneline”>本篇博客参考：http://blog.cloudera.com/blog/2015/03/how-to-tune-your-apache-spark-jobs-part-2/ 软件版本：CDH：5.7…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/fansy1990″>来自： <span class=”blog_title”> fansy1990的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/SQLDebug_Fan/article/details/7881793,BlogCommendHotData_0,index_50"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/SQLDebug_Fan/article/details/7881793,BlogCommendHotData_0,index_50"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/SQLDebug_Fan/article/details/7881793″ target=”_blank” title=”DELPHI高性能大容量SOCKET并发（一）：IOCP完成端口例子介绍”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
DELPHI高性能大容量SOCKET并发（一）：IOCP完成端口例子介绍 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>08-18</span>
<span class=”read-num hover-hide”>
阅读数
3万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/SQLDebug_Fan/article/details/7881793″ target=”_blank” title=”DELPHI高性能大容量SOCKET并发（一）：IOCP完成端口例子介绍”>
<span class=”desc oneline”>例子主要包括IOCP控件封装、服务端实现、传输协议和日志、控制、SQL查询、上传、下载等协议实现，并包括一些初步的性能测试结果。
服务端：
界面截图如下：
提供服务和桌面方式运行，桌面方式可直接打…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/SQLDebug_Fan”>来自： <span class=”blog_title”> SQLDebug_Fan的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/Yan_Chou/article/details/53406095,BlogCommendHotData_1,index_51"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/Yan_Chou/article/details/53406095,BlogCommendHotData_1,index_51"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Yan_Chou/article/details/53406095″ target=”_blank” title=”frp配置本地服务端口到服务器80端口”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
frp配置本地服务端口到服务器80端口 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-30</span>
<span class=”read-num hover-hide”>
阅读数
1万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Yan_Chou/article/details/53406095″ target=”_blank” title=”frp配置本地服务端口到服务器80端口”>
<span class=”desc oneline”>搭建环境：
ubuntu 16.04 LTS （本地服务计算机） ubuntu 14.04 LTS(阿里云服务器)
apache tomcat 7
java 7
frp 0.8.1 linux
搭建…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/Yan_Chou”>来自： <span class=”blog_title”> Anteoy的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/opensure/article/details/46453681,BlogCommendHotData_2,index_52"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/opensure/article/details/46453681,BlogCommendHotData_2,index_52"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/opensure/article/details/46453681″ target=”_blank” title=”Hbase 布隆过滤器BloomFilter介绍”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Hbase 布隆过滤器BloomFilter介绍 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-11</span>
<span class=”read-num hover-hide”>
阅读数
1万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/opensure/article/details/46453681″ target=”_blank” title=”Hbase 布隆过滤器BloomFilter介绍”>
<span class=”desc oneline”>1、主要功能
提高随机读的性能
2、存储开销
bloom filter的数据存在StoreFile的meta中，一旦写入无法更新，因为StoreFile是不可变的。Bloomfilter是…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/opensure”>来自： <span class=”blog_title”> opensure的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/diandianxiyu/article/details/53068012,BlogCommendHotData_3,index_53"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/diandianxiyu/article/details/53068012,BlogCommendHotData_3,index_53"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/diandianxiyu/article/details/53068012″ target=”_blank” title=”【小程序】微信小程序开发实践”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
【小程序】微信小程序开发实践 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-07</span>
<span class=”read-num hover-hide”>
阅读数
13万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/diandianxiyu/article/details/53068012″ target=”_blank” title=”【小程序】微信小程序开发实践”>
<span class=”desc oneline”>帐号相关流程注册范围
企业
政府
媒体
其他组织换句话讲就是不让个人开发者注册。 :)填写企业信息不能使用和之前的公众号账户相同的邮箱,也就是说小程序是和微信公众号一个层级的。填写公司机构信息,对公账…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/diandianxiyu”>来自： <span class=”blog_title”> 小雨同学的技术博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/KingWTD/article/details/54425903,BlogCommendHotData_4,index_54"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/KingWTD/article/details/54425903,BlogCommendHotData_4,index_54"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/KingWTD/article/details/54425903″ target=”_blank” title=”Chrome浏览器 – 抢票插件的制作与插件工具（上）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Chrome浏览器 – 抢票插件的制作与插件工具（上） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-14</span>
<span class=”read-num hover-hide”>
阅读数
6967</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/KingWTD/article/details/54425903″ target=”_blank” title=”Chrome浏览器 – 抢票插件的制作与插件工具（上）”>
<span class=”desc oneline”>关于Chrome浏览器插件的博客有很多，它的应用场景也很多。一般来讲，Chrome浏览器插件分为三大类：ContentScript、Background和Popup，今天来说一下我们都比较关注的一个插…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/KingWTD”>来自： <span class=”blog_title”> KingWTD的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/leishenop/article/details/62246975,BlogCommendHotData_5,index_55"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/leishenop/article/details/62246975,BlogCommendHotData_5,index_55"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/leishenop/article/details/62246975″ target=”_blank” title=”腾讯云的一道面试题—– 一致性Hash算法”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
腾讯云的一道面试题—– 一致性Hash算法 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-15</span>
<span class=”read-num hover-hide”>
阅读数
3771</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/leishenop/article/details/62246975″ target=”_blank” title=”腾讯云的一道面试题—– 一致性Hash算法”>
<span class=”desc oneline”>今天在面试腾讯的时候，被面试官问到这块的知识，之前只是了解过，没有真正深入的了解，现在才知道自己的差距，要更加努力的补充这些技术知识了。下面是非常好的这块的资料，和大家一起分享。
一致性 hash …</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/leishenop”>来自： <span class=”blog_title”> leishenop的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_36892341/article/details/73918672,BlogCommendHotData_6,index_56"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_36892341/article/details/73918672,BlogCommendHotData_6,index_56"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36892341/article/details/73918672″ target=”_blank” title=”linux上安装Docker(非常简单的安装方法)”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
linux上安装Docker(非常简单的安装方法) </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-29</span>
<span class=”read-num hover-hide”>
阅读数
14万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_36892341/article/details/73918672″ target=”_blank” title=”linux上安装Docker(非常简单的安装方法)”>
<span class=”desc oneline”>最近比较有空，大四出来实习几个月了，作为实习狗的我，被叫去研究Docker了，汗汗！
Docker的三大核心概念：镜像、容器、仓库
镜像：类似虚拟机的镜像、用俗话说就是安装文件。
容器：类似一个轻量…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_36892341″>来自： <span class=”blog_title”> 我走小路的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/crazy1235/article/details/42614603,BlogCommendHotData_7,index_57"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/crazy1235/article/details/42614603,BlogCommendHotData_7,index_57"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/crazy1235/article/details/42614603″ target=”_blank” title=”Android百度地图开发（一）之初体验”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Android百度地图开发（一）之初体验 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-11</span>
<span class=”read-num hover-hide”>
阅读数
11万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/crazy1235/article/details/42614603″ target=”_blank” title=”Android百度地图开发（一）之初体验”>
<span class=”desc oneline”>转载请注明出处：
做关于位置或者定位的app的时候免不了使用地图功能，本人最近由于项目的需求需要使用百度地图的一些功能，所以这几天研究了一下，现写一下blog记录一下，欢迎大家评论指正！
一…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/crazy1235″>来自： <span class=”blog_title”> crazy_jack</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_41410358/article/details/79679130,BlogCommendHotData_8,index_58"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_41410358/article/details/79679130,BlogCommendHotData_8,index_58"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_41410358/article/details/79679130″ target=”_blank” title=”myEclipse10安装以及破解”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
myEclipse10安装以及破解 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-24</span>
<span class=”read-num hover-hide”>
阅读数
5万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_41410358/article/details/79679130″ target=”_blank” title=”myEclipse10安装以及破解”>
<span class=”desc oneline”>myEclipse10可以去很多地方下载，我这里是从电脑管家下载的，下载完成后安装，安装很简单，不在多说安装完成后会发现你只有一个月的试用时间，我们就需要破解，这里需要下载一个破解补丁https://…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_41410358″>来自： <span class=”blog_title”> 拥之则安的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/u011860731/article/details/48733073,BlogCommendHotData_9,index_59"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/u011860731/article/details/48733073,BlogCommendHotData_9,index_59"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u011860731/article/details/48733073″ target=”_blank” title=”ThreadLocal的设计理念与作用”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
ThreadLocal的设计理念与作用 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-25</span>
<span class=”read-num hover-hide”>
阅读数
2万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/u011860731/article/details/48733073″ target=”_blank” title=”ThreadLocal的设计理念与作用”>
<span class=”desc oneline”>Java中的ThreadLocal类允许我们创建只能被同一个线程读写的变量。因此，如果一段代码含有一个ThreadLocal变量的引用，即使两个线程同时执行这段代码，它们也无法访问到对方的Thread…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/u011860731″>来自： <span class=”blog_title”> u011860731的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/wuchengzeng/article/details/50037611,BlogCommendHotData_10,index_60"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/wuchengzeng/article/details/50037611,BlogCommendHotData_10,index_60"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/wuchengzeng/article/details/50037611″ target=”_blank” title=”jquery/js实现一个网页同时调用多个倒计时(最新的)”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
jquery/js实现一个网页同时调用多个倒计时(最新的) </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-25</span>
<span class=”read-num hover-hide”>
阅读数
21万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/wuchengzeng/article/details/50037611″ target=”_blank” title=”jquery/js实现一个网页同时调用多个倒计时(最新的)”>
<span class=”desc oneline”>jquery/js实现一个网页同时调用多个倒计时(最新的)
最近需要网页添加多个倒计时. 查阅网络,基本上都是千遍一律的不好用. 自己按需写了个.希望对大家有用. 有用请赞一个哦!
//js
…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/wuchengzeng”>来自： <span class=”blog_title”> websites</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/ohmygirl/article/details/17846199,BlogCommendHotData_11,index_61"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/ohmygirl/article/details/17846199,BlogCommendHotData_11,index_61"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/ohmygirl/article/details/17846199″ target=”_blank” title=”【HTTP】Fiddler（一） – Fiddler简介”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
【HTTP】Fiddler（一） – Fiddler简介 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>01-04</span>
<span class=”read-num hover-hide”>
阅读数
29万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/ohmygirl/article/details/17846199″ target=”_blank” title=”【HTTP】Fiddler（一） – Fiddler简介”>
<span class=”desc oneline”>1.为什么是Fiddler?
抓包工具有很多，小到最常用的web调试工具firebug，达到通用的强大的抓包工具wireshark.为什么使用fiddler?原因如下：
a.Firebug虽然可以抓包…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/ohmygirl”>来自： <span class=”blog_title”> 专注、专心</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/arlanhon/article/details/18908961,BlogCommendHotData_12,index_62"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/arlanhon/article/details/18908961,BlogCommendHotData_12,index_62"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/arlanhon/article/details/18908961″ target=”_blank” title=”在win7系统硬盘安装Fedora20″>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
在win7系统硬盘安装Fedora20 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>02-03</span>
<span class=”read-num hover-hide”>
阅读数
9178</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/arlanhon/article/details/18908961″ target=”_blank” title=”在win7系统硬盘安装Fedora20″>
<span class=”desc oneline”>一、实验目的：
在linux系统中塔建一个用eclipse开发的环境（本人的电脑配置一盘，故没打算采用虚拟机的形式），经过网上的一些资料，最终选反了Fedora20版本的，其实可以用U…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/arlanhon”>来自： <span class=”blog_title”> 汐 朔</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/xiaowei_cqu/article/details/26471527,BlogCommendHotData_13,index_63"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/xiaowei_cqu/article/details/26471527,BlogCommendHotData_13,index_63"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xiaowei_cqu/article/details/26471527″ target=”_blank” title=”【图像处理】透视变换 Perspective Transformation”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
【图像处理】透视变换 Perspective Transformation </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>05-26</span>
<span class=”read-num hover-hide”>
阅读数
11万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/xiaowei_cqu/article/details/26471527″ target=”_blank” title=”【图像处理】透视变换 Perspective Transformation”>
<span class=”desc oneline”>透视变换(Perspective Transformation)是将图片投影到一个新的视平面(Viewing Plane)，也称作投影映射(Projective Mapping)。通用的变换公式为：u…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/xiaowei_cqu”>来自： <span class=”blog_title”> 小魏的修行路</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/StillCity/article/details/16988317,BlogCommendHotData_14,index_64"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/StillCity/article/details/16988317,BlogCommendHotData_14,index_64"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/StillCity/article/details/16988317″ target=”_blank” title=”openfire 3.8.2 源码部署 /开发配置 / 二次开发”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
openfire 3.8.2 源码部署 /开发配置 / 二次开发 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-27</span>
<span class=”read-num hover-hide”>
阅读数
4534</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/StillCity/article/details/16988317″ target=”_blank” title=”openfire 3.8.2 源码部署 /开发配置 / 二次开发”>
<span class=”desc oneline”>最近新搞了openfire 从网上找了很多源码部署的相关文章但都是大同小异，拷贝加修改，我如是按照各个文章版本部署目前最新的3.8.2版本，无一例外，各种报错，头疼死我也，一次次失败，我TMD就想为啥…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/StillCity”>来自： <span class=”blog_title”> StillCity的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/jnulzl/article/details/49894041,BlogCommendHotData_15,index_65"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/jnulzl/article/details/49894041,BlogCommendHotData_15,index_65"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/jnulzl/article/details/49894041″ target=”_blank” title=”史上最好的LDA(线性判别分析)教程”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
史上最好的LDA(线性判别分析)教程 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-17</span>
<span class=”read-num hover-hide”>
阅读数
2万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/jnulzl/article/details/49894041″ target=”_blank” title=”史上最好的LDA(线性判别分析)教程”>
<span class=”desc oneline”>一、前言最近由于研究需要，要用到线性判别分析(LDA)。于是找了很多资料来看，结果发现大部分讲的都是理论知识，因此最后还是看的一知半解，后来终于找到了个英文的文档，作者由PCA引入LDA，看过后豁然开…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/jnulzl”>来自： <span class=”blog_title”> jnulzl的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_22073849/article/details/71809243,BlogCommendHotData_16,index_66"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/qq_22073849/article/details/71809243,BlogCommendHotData_16,index_66"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_22073849/article/details/71809243″ target=”_blank” title=”执行转换时如何让Kettle记录错误并继续执行？——记一种解决方案”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
执行转换时如何让Kettle记录错误并继续执行？——记一种解决方案 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>05-13</span>
<span class=”read-num hover-hide”>
阅读数
1万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/qq_22073849/article/details/71809243″ target=”_blank” title=”执行转换时如何让Kettle记录错误并继续执行？——记一种解决方案”>
<span class=”desc oneline”>如题，近几天在利用Kettle进行数据迁移的工作（也就是把数据全量导入到新数据库中，其中有些字段有些许调整），在写好脚本并执行后发现一个很严重的问题——每次脚本在执行到某个位置的时候就会意外停止，如下…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/qq_22073849″>来自： <span class=”blog_title”> Amour</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/ymj7150697/article/details/7384126,BlogCommendHotData_17,index_67"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/ymj7150697/article/details/7384126,BlogCommendHotData_17,index_67"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/ymj7150697/article/details/7384126″ target=”_blank” title=”expat介绍文档翻译”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
expat介绍文档翻译 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>03-22</span>
<span class=”read-num hover-hide”>
阅读数
2万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/ymj7150697/article/details/7384126″ target=”_blank” title=”expat介绍文档翻译”>
<span class=”desc oneline”>原文地址：http://www.xml.com/pub/a/1999/09/expat/index.html
因为需要用，所以才翻译了这个文档。但总归赖于英语水平很有限，翻译出来的中文有可能…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/ymj7150697″>来自： <span class=”blog_title”> ymj7150697的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/mingming24/article/details/46595127,BlogCommendHotData_18,index_68"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/mingming24/article/details/46595127,BlogCommendHotData_18,index_68"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/mingming24/article/details/46595127″ target=”_blank” title=”改造动态framework为静态framework”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
改造动态framework为静态framework </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>06-22</span>
<span class=”read-num hover-hide”>
阅读数
1097</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/mingming24/article/details/46595127″ target=”_blank” title=”改造动态framework为静态framework”>
<span class=”desc oneline”>公司产品之前使用xmpp作为底层库，之前同事编译自己的sdk静态库想生成.a库，但是各种编译问题（其实耐心修改配置都能解决），但是从百度找到方案用framework可以解决，所以最终使用的是frame…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/mingming24″>来自： <span class=”blog_title”> mingming24的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/silentpebble/article/details/41279285,BlogCommendHotData_19,index_69"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/silentpebble/article/details/41279285,BlogCommendHotData_19,index_69"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/silentpebble/article/details/41279285″ target=”_blank” title=”centos 查看命令源码”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
centos 查看命令源码 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>11-19</span>
<span class=”read-num hover-hide”>
阅读数
3万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/silentpebble/article/details/41279285″ target=”_blank” title=”centos 查看命令源码”>
<span class=”desc oneline”># yum install yum-utils
设置源:
[base-src]
name=CentOS-5.4 – Base src –
baseurl=http://vault.ce…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/silentpebble”>来自： <span class=”blog_title”> linux/unix</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/feimengjuan/article/details/69666981,BlogCommendHotData_20,index_70"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/feimengjuan/article/details/69666981,BlogCommendHotData_20,index_70"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/feimengjuan/article/details/69666981″ target=”_blank” title=”你应该知道的9篇深度学习论文（CNNs 理解）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
你应该知道的9篇深度学习论文（CNNs 理解） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>04-08</span>
<span class=”read-num hover-hide”>
阅读数
3万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/feimengjuan/article/details/69666981″ target=”_blank” title=”你应该知道的9篇深度学习论文（CNNs 理解）”>
<span class=”desc oneline”>当时看到英文的博客，本想翻译给感兴趣的同学们看看，没想到已经有人翻译，于是作个转载，留给自己和更多的人学习。
英文博客：https://adeshpande3.github.io/adeshpan…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/feimengjuan”>来自： <span class=”blog_title”> 以后以后的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/jackfrued/article/details/44921941,BlogCommendHotData_21,index_71"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/jackfrued/article/details/44921941,BlogCommendHotData_21,index_71"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/jackfrued/article/details/44921941″ target=”_blank” title=”Java面试题全集（上）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Java面试题全集（上） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>04-08</span>
<span class=”read-num hover-hide”>
阅读数
92万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/jackfrued/article/details/44921941″ target=”_blank” title=”Java面试题全集（上）”>
<span class=”desc oneline”>2013年年底的时候，我看到了网上流传的一个叫做《Java面试题大全》的东西，认真的阅读了以后发现里面的很多题目是重复且没有价值的题目，还有不少的参考答案也是错误的，于是我花了半个月时间对这个所谓的《…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/jackfrued”>来自： <span class=”blog_title”> 骆昊的技术专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/gefangshuai/article/details/50328451,BlogCommendHotData_22,index_72"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/gefangshuai/article/details/50328451,BlogCommendHotData_22,index_72"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/gefangshuai/article/details/50328451″ target=”_blank” title=”关于SpringBoot bean无法注入的问题（与文件包位置有关）”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
关于SpringBoot bean无法注入的问题（与文件包位置有关） </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>12-16</span>
<span class=”read-num hover-hide”>
阅读数
10万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/gefangshuai/article/details/50328451″ target=”_blank” title=”关于SpringBoot bean无法注入的问题（与文件包位置有关）”>
<span class=”desc oneline”>问题场景描述整个项目通过Maven构建，大致结构如下：
核心Spring框架一个module spring-boot-base
service和dao一个module server-core
提供系统…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/gefangshuai”>来自： <span class=”blog_title”> 开发随笔</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/Tiaaaaa/article/details/58116346,BlogCommendHotData_23,index_73"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/Tiaaaaa/article/details/58116346,BlogCommendHotData_23,index_73"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Tiaaaaa/article/details/58116346″ target=”_blank” title=”R语言逻辑回归、ROC曲线和十折交叉验证”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
R语言逻辑回归、ROC曲线和十折交叉验证 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>02-27</span>
<span class=”read-num hover-hide”>
阅读数
3万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/Tiaaaaa/article/details/58116346″ target=”_blank” title=”R语言逻辑回归、ROC曲线和十折交叉验证”>
<span class=”desc oneline”>自己整理编写的逻辑回归模板，作为学习笔记记录分享。数据集用的是14个自变量Xi，一个因变量Y的australian数据集。
1. 测试集和训练集3、7分组
australian …</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/Tiaaaaa”>来自： <span class=”blog_title”> Tiaaaaa的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/beijiu5854/article/details/77897767,BlogCommendHotData_24,index_74"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/beijiu5854/article/details/77897767,BlogCommendHotData_24,index_74"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/beijiu5854/article/details/77897767″ target=”_blank” title=”Ubuntu16.04下完美切换Python版本”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Ubuntu16.04下完美切换Python版本 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-08</span>
<span class=”read-num hover-hide”>
阅读数
2万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/beijiu5854/article/details/77897767″ target=”_blank” title=”Ubuntu16.04下完美切换Python版本”>
<span class=”desc oneline”>转载自http://blog.csdn.net/u013894834/article/details/75305752
Ubuntu16.04下完美切换Python版本（亲测）
对于ubuntu …</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/beijiu5854″>来自： <span class=”blog_title”> beijiu5854的博客</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/irving512/article/details/78119192,BlogCommendHotData_25,index_75"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/irving512/article/details/78119192,BlogCommendHotData_25,index_75"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/irving512/article/details/78119192″ target=”_blank” title=”Python(2) 基础语法”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
Python(2) 基础语法 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>09-28</span>
<span class=”read-num hover-hide”>
阅读数
4004</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/irving512/article/details/78119192″ target=”_blank” title=”Python(2) 基础语法”>
<span class=”desc oneline”>1. 模块1.1. 从某模块导入函数import somemodule
from somemodule import somefunction
from somemodule import somef…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/irving512″>来自： <span class=”blog_title”> 清欢</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box recommend-box-ident type_blog clearfix” data-track-view=”{"mod":"popu_614","con":",https://blog.csdn.net/VBcom/article/details/7245186,BlogCommendHotData_26,index_76"}” data-track-click=”{"mod":"popu_614","con":",https://blog.csdn.net/VBcom/article/details/7245186,BlogCommendHotData_26,index_76"}” data-flg=”true”>
<div class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/VBcom/article/details/7245186″ target=”_blank” title=”DirectX修复工具增强版”>
<h4 class=”text-truncate oneline” style=”width: 802px;”>
DirectX修复工具增强版 </h4>
<div class=”info-box d-flex align-content-center”>
<p class=”date-and-readNum oneline”>
<span class=”date hover-show”>02-09</span>
<span class=”read-num hover-hide”>
阅读数
183万+</span>
</p>
</div>
</a>
<p class=”content” style=”width: 962px;”>
<a href=”https://blog.csdn.net/VBcom/article/details/7245186″ target=”_blank” title=”DirectX修复工具增强版”>
<span class=”desc oneline”>最后更新：2018-12-20
DirectX修复工具最新版：DirectX Repair V3.8 增强版 NEW!
版本号：V3.8.0.11638
大小: 107MB/7z格式压缩，18…</span>
</a>
<span class=”blog_title_box oneline “>
<span class=”type-show type-show-blog type-show-after”>博文</span>
<a target=”_blank” href=”https://blog.csdn.net/VBcom”>来自： <span class=”blog_title”> VBcom的专栏</span></a>
</span>
</p>
</div>
</div>
<div class=”recommend-item-box type_hot_word”>
<div class=”content clearfix” style=”width: 962px;”>
<div class=”word float-left”>
<span>
<a href=”https://edu.csdn.net/courses/o317_s318_k ” target=”_blank”>
Hadoop </a></span>
<span>
<a href=”https://edu.csdn.net/courses/o317_s318_k ” target=”_blank”>
Hadoop培训 </a></span>
<span>
<a href=”https://edu.csdn.net/combos/o317_s318_l0_t ” target=”_blank”>
Hadoop学习 </a></span>
<span>
<a href=”https://edu.csdn.net/combos/o317_s318_l0_t ” target=”_blank”>
Hadoop课程 </a></span>
<span>
<a href=”https://edu.csdn.net/combos/o317_s318_l0_t ” target=”_blank”>
Hadoop视频教程 </a></span>
</div>
</div>
<div class=”content clearfix” style=”width: 962px;”>
<div class=”float-left”>
<span>
<a href=”https://www.csdn.net/gather_20/MtjaMgzsNTktYmxvZwO0O0OO0O0O.html” target=”_blank”>
c# 内存 获取cpu</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_24/MtTakg5sMDU3NS1ibG9n.html” target=”_blank”>
android cpu显示 内存</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_2d/NtzaQg0sOTgtYmxvZwO0O0OO0O0O.html” target=”_blank”>
c++ 函数 查询cpu和内存</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_2a/NtzaMg5sNS1ibG9n.html” target=”_blank”>
c# 性能调优工具</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_22/MtzaQg5sMjEtYmxvZwO0O0OO0O0O.html” target=”_blank”>
c++ hadoop 编译环境</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_4a/Mtzagg2sMS1lZHUO0O0O.html” target=”_blank”>
python配置图文教程</a>
</span>
<span>
<a href=”https://www.csdn.net/gather_4a/MtTaQgxsMC1lZHUO0O0O.html” target=”_blank”>
python内存挂教程</a>
</span>
</div>
</div>
</div>
<div class=”recommend-loading-box”>
<img src=”https://csdnimg.cn/release/phoenix/images/feedLoading.gif”>
</div>
<div class=”recommend-end-box”>
<p class=”text-center”>没有更多推荐了，<a href=”https://blog.csdn.net/” class=”c-blue c-blue-hover c-blue-focus”>返回首页</a></p>
</div>
</div>
</main>