2、Yarn集群wordcount测试_Hadoop性能测试_托管Hadoop

帮助文档（华北一、二）

使用BigDataBench的数据源，从10GB开始，每次增加10GB，直至200GB。生成数据源代码如下：

● yarn-wordcount数据源生成代码

wget

http://prof.ict.ac.cn/bdb_uploads/bdb_3_1/packages/BigDataBench_V3.2.1_Hadoop.tar.gz

tar zxf BigDataBench_V3.2.1_Hadoop.tar.gz

cd BigDataBench_V3.2.1_Hadoop_Hive/MicroBenchmarks sh genData_MicroBenchmarks.sh 200

依次从数据源取数据，进行wordcount测试，并打印结果，代码如下：

● yarn-wordcount-run.sh

#!/bin/bash

source /home/hadoop/.bashrc

WORK_DIR=`pwd`

echo "WORK_DIR=$WORK_DIR data should be put in $WORK_DIR/data-MicroBenchmarks/in"

${HADOOP_HOME}/bin/hadoop fs -mkdir -p /MicroBenchmarks/in/

${HADOOP_HOME}/bin/hadoop fs -mv /MicroBenchmarks/in/* ${WORK_DIR}/data-MicroBenchmarks/in/

for i in {1..20}

#move 10GB file from ${WORK_DIR}/data-MicroBenchmarks/in/ to /MicroBenchmarks/in/

for f in `${HADOOP_HOME}/bin/hadoop fs -ls ${WORK_DIR}/data-

MicroBenchmarks/in/ | grep lda_wiki1w | awk '{print $NF}' | head -n 20`

${HADOOP_HOME}/bin/hadoop fs -mv $f /MicroBenchmarks/in/

Done

${HADOOP_HOME}/bin/hadoop fs -rmr ${WORK_DIR}/data-MicroBenchmarks/out/wordcount

echo -n "file count = "

${HADOOP_HOME}/bin/hadoop fs -ls /MicroBenchmarks/in/ | grep lda_wiki1w | wc -l

time ${HADOOP_HOME}/bin/hadoop jar

${HADOOP_HOME}/share/hadoop/mapreduce/hadoop-mapreduce-examples-*.jar

wordcount /MicroBenchmarks/in/ ${WORK_DIR}/data-MicroBenchmarks/out/wordcount

done

对四个集群的测试结果如下：

● 表2.1 yarn集群wordcount测试结果：

	耗时（s）				处理速度（MB/s）
	B2	B2-x	D1	D1-x	B2	B2-x	D1	D1-x
10G	332	176	319	177	30.84	58.18	32.10	57.85
20G	624	319	618	322	32.82	64.20	33.14	63.60
30G	990	455	922	462	31.03	67.52	33.32	66.49
40G	1267	599	1210	613	32.33	68.38	33.85	66.82
50G	1536	752	1518	757	33.33	68.09	33.73	67.64
60G	1834	903	1840	925	33.50	68.04	33.39	66.42
70G	2113	1041	2111	1064	33.92	68.86	33.96	67.37
80G	2504	1184	2411	1215	32.72	69.19	33.98	67.42
90G	2785	1334	2746	1360	33.09	69.09	33.56	67.76
100G	3073	1476	3006	1469	33.32	69.38	34.07	69.71
110G	3290	1624	3312	1625	34.24	69.36	34.01	69.32
120G	3633	1797	3619	1793	33.82	68.38	33.95	68.53
130G	3986	1910	3915	1930	33.40	69.70	34.00	68.97
140G	4200	2053	4218	2090	34.13	69.83	33.99	68.59
150G	4562	2208	4517	2226	33.67	69.57	34.00	69.00
160G	4813	2356	4732	2369	34.04	69.54	33.91	69.16
170G	5146	2496	5120	2521	33.83	69.74	34.00	69.05
180G	5453	2686	5413	2662	33.80	68.62	34.05	69.24
190G	5712	2795	5723	2804	34.06	69.61	34.00	69.39
200G	6147	2938	6031	2953	33.32	69.71	33.96	69.35

● 图2.1 yarn集群wordcount处理耗时

● 图2.2 yarn集群wordcount处理速度

BigDataBench介绍内容与下载链接：

http://prof.ict.ac.cn/BigDataBench/dowloads/

最新活动

产品

解决方案

政务云

工业互联网

云市场

支持与服务

合作伙伴

浪潮云计算

支持与服务

合作伙伴生态

关于浪潮云

特色平台通道

文档