Cassandra压力测试

  • cassandra-stress
    • 介绍
    • 单节点测试
      • 写入测试
      • 采用cql3协议写入
      • 读取测试
      • 没有预热读取
      • 给定一段时间测试最大可以读取多少条记录
      • 混合模式读大于写
      • 混合模式写大于读
    • 自定义schema
      • 示例
      • 只有写入
      • 读写混合模式
      • 只有读取
    • G1测试
    • 集群测试
  • YCSB

cassandra-stress

机器配置:SSD,RAID0,62G内存,24核CPU
默认JVM配置:-Xms8192M -Xmx8192M -Xmn2048M
C版本:2.2.6,–> 3.6

说明:Cassandra对于内存的计算方式是如果超过8G,则堆的最大内存只能配置8G,而新生代的内存为堆内存的1/4=2G

准备测试目录,修改配置文件的directory选项,把localhost相关的地址改成实际地址,便于后面集群模式(确保集群seeds一致):

节点数量 种子节点 集群名称 自定义配置文件
单节点 10.21.21.10 TestBenchmark1 blogpost1.yaml
四节点 10.21.21.130 TestBenchmark blogpost4.yaml

为什么单节点和多节点采用不同的集群名称,因为不止测试一次,可能要更改GC配置,再次测试,如果只用一个集群名称,从一个节点增加到四个节点后,如果要再测试一个节点的就比较麻烦。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
wget http://192.168.47.211:8000/apache-cassandra-2.2.6-bin.tar.gz && tar zxf apache-cassandra-2.2.6-bin.tar.gz

cluster=TestBenchmark1 #TestBenchmark
seeds=10.21.21.10 #10.21.21.130

mkdir test
host=`ifconfig | grep "10.21.21." | awk '/inet addr/{sub("addr:",""); print $2}'`
sed -i -e "s/localhost/$host/g" apache-cassandra-2.2.6/conf/cassandra.yaml
sed -i -e "s#localhost#$host#g" apache-cassandra-2.2.6/conf/cassandra-env.sh
sed -i -e "s/127.0.0.1/$seeds/g" apache-cassandra-2.2.6/conf/cassandra.yaml
sed -i -e "s/Test Cluster/$cluster/g" apache-cassandra-2.2.6/conf/cassandra.yaml
echo "data_file_directories: [/home/admin/test/data]" >> apache-cassandra-2.2.6/conf/cassandra.yaml
echo "saved_caches_directory: /home/admin/test/saved_caches" >> apache-cassandra-2.2.6/conf/cassandra.yaml
echo "commitlog_directory: /home/admin/test/commitlog" >> apache-cassandra-2.2.6/conf/cassandra.yaml
echo "auto_bootstrap: false" >> apache-cassandra-2.2.6/conf/cassandra.yaml

为了测试的准确性,不要在按照有Cassandra的当前节点做压力测试,可以在集群之外安装一个Cassandra,
但是注意不要启动他,使用cassandra-stress的-node连接到目标集群也可以测试。

介绍

Cassandra从很早的版本就自带了cassandra-stress压力测试工具,它的使用方法在cassandra-stress后添加命令和选项。
其中常用的命令一般只用到:write、read、mixed、user。其中单纯的write和read只测试读和写,mixed则测试同时读写。
user是2.1之后新增的,通过自定义配置文件,在配置文件中可以指定insert和query查询语句。
命令command没有以-开头,而选项[options]有两种方式:-选项名称 选项值或者选项名称=选项值

1
2
cd apache-cassandra-2.2.6/tools/bin/
cassandra-stress command [options]

常用的选项有

  1. 线程数量:-rate threads=? [limit=?],比如-rate threads=100
  2. 表结构设置-schema [replication(?)] [keyspace=?] [compaction(?)] [compression=?]
  3. 日志设置:-log level=verbose
  4. 运行模式的几种示例,因为Cassandra本身提供了多种交互方式比如thrift协议,CQL协议等
    1
    2
    3
    -mode thrift [smart]
    -mode native [unprepared] cql3 [compression=?]
    -mode simplenative [prepared] cql3

在2.1.5之后,添加了一些选项(通常针对user命令)

  1. profile=?:指定YAML配置文件,需要自己编写DML,插入,查询
  2. ops(?):指定操作类型和数量,比如ops(inserts=1),或者ops(queries=2),其中queries需要用指定的查询名称代替
  3. n=?:指定操作数量,比如要写入10万条数据,n=100000; 要读取1000条数据,n=1000
  4. truncate=?,是否需要清空表,可选项有:never(默认值),one,always
  5. cl=?:一致性级别,可选项有:ONE,QUORUM,LOCAL_QUORUM,EACH_QUORUM,ALL,ANY,LOCAL_ONE(默认值)

注意选项名称1=选项值必须放在-选项名称2 选项值前面,如果把选项名称1放在后面,会被认为作为选项名称2的子选项。
比如正确的用法:truncate=one -node xxx,而不能是:-node xxx truncate=one,后者会认为是-node的子选项。

一个示例:写入100万条数据,一致性级别为Local_Quorum,客户端线程数=500个,32个列,副本数据=3个

1
2
cassandra-stress write n=1000000 cl=LOCAL_QUORUM -rate threads=500 \
-col "size=fixed(2048)" "n=fixed(32)" -schema "replication(factor=3)" -node 192.168.10.12

单节点测试

一键测试命令:nohup sh stress1.sh > stress1.log 2>&1 &

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
/home/admin/apache-cassandra-2.2.6/bin/nodetool -h 10.21.21.10 status
cd /home/admin/apache-cassandra-2.2.6/tools/bin
#标准测试
########
./cassandra-stress write n=1000000 -node 10.21.21.10 -log file=SINGLENODE1_write_1M.log
./cassandra-stress read n=200000 -node 10.21.21.10 -log file=SINGLENODE2_read_2.log
./cassandra-stress read n=200000 no-warmup -node 10.21.21.10 -log file=SINGLENODE3_read_2_nowarm.log
./cassandra-stress read duration=3m -node 10.21.21.10 -log file=SINGLENODE4_read_3min.log

#混合模式
########
#./cassandra-stress write n=1 cl=one -mode native cql3 -log file=SINGLENODE_create_schema.log -node 10.21.21.10
#./cassandra-stress write n=1000000 cl=one -mode native cql3 -schema keyspace="keyspace1" -log file=SINGLENODE5_write_1M_rows.log -node 10.21.21.10

./cassandra-stress mixed ratio\(write=1,read=3\) n=100000 cl=ONE -pop dist=UNIFORM\(1..1000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 threads\<=256 -log file=SINGLENODE5_mixed_write1read3_1M.log -node 10.21.21.10
./cassandra-stress mixed ratio\(write=3,read=1\) n=100000 cl=ONE -pop dist=UNIFORM\(1..1000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 threads\<=256 -log file=SINGLENODE6_mixed_write3read1_1M.log -node 10.21.21.10

#./cassandra-stress write n=1000000 cl=one -mode native cql3 -schema keyspace="keyspace1" -pop seq=1..1000000 -node 10.21.21.10
#./cassandra-stress read n=1000000 cl=one -mode native cql3 -schema keyspace="keyspace1" -pop seq=1..1000000 -node 10.21.21.10

#自定义表
########
./cassandra-stress user profile=./blogpost1.yaml ops\(insert=1\) -node 10.21.21.10 -log file=SINGLENODE7_blog_insert.log
./cassandra-stress user profile=./blogpost1.yaml ops\(singlepost=2,timeline=1,insert=1\) -node 10.21.21.10 -log file=SINGLENODE8_blog_mixed.log
./cassandra-stress user profile=./blogpost1.yaml ops\(singlepost=1\) -node 10.21.21.10 -log file=SINGLENODE9_blog_q1.log
./cassandra-stress user profile=./blogpost1.yaml ops\(timeline=1\) -node 10.21.21.10 -log file=SINGLENODE10_blog_q2.log

#支持图形显示结果(3.2以上支持)
########
#cassandra-stress write n=100000 -rate threads=10 -graph file=example-benchmark.html title=example revision=benchmark-0
#cassandra-stress mixed n=100000 -rate threads=10 -graph file=example-benchmark.html title=example revision=benchmark-0

写入测试

测试100万条数据的写入速度:./cassandra-stress write n=1000000 -node 10.21.21.10 -log file=SINGLENODE1_write_1M.log
测试结果:每秒钟写入5.8万条,由于每个row key都是唯一的,所以100万条数据有100万个分区,每秒钟的写入量=每秒钟写入的分区数=每秒钟写入的行数。
每条记录的写延迟平均3.4ms,写延迟小于4.1ms的占比95,写延迟小于21ms的占比99%,也就是说99%的记录写延迟在21ms以下。
最大一条记录的写延迟为180ms。GC一共发生了5次,GC的大小为7G,GC时间花了1秒,平均GC时间为143ms。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Results: [write n=1000000], 200 threads
op rate : 57901 [WRITE:57901]
partition rate : 57901 [WRITE:57901]
row rate : 57901 [WRITE:57901]
latency mean : 3.4 [WRITE:3ps.4]
latency median : 3.2 [WRITE:3.2]
latency 95th percentile : 4.1 [WRITE:4.1]
latency 99th percentile : 5.1 [WRITE:5.1]
latency 99.9th percentile : 21.3 [WRITE:21.3]
latency max : 180.1 [WRITE:180.1]
Total partitions : 1000000 [WRITE:1000000]
Total errors : 0 [WRITE:0]
total gc count : 5
total gc mb : 7362
total gc time (s) : 1
avg gc time(ms) : 143
stdev gc time(ms) : 42
Total operation time : 00:00:17

客户端在集群之外测试时,每秒钟的写入量增大到11万。不过GC大小和时间没有多大变化,当然因为每秒钟写入量增多,总共10万条数据,花费的时间也就不同了。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 114509 [WRITE:114509]
partition rate : 114509 [WRITE:114509]
row rate : 114509 [WRITE:114509]
latency mean : 1.7 [WRITE:1.7]
latency median : 1.4 [WRITE:1.4]
latency 95th percentile : 2.0 [WRITE:2.0]
latency 99th percentile : 3.0 [WRITE:3.0]
latency 99.9th percentile : 110.1 [WRITE:110.1]
latency max : 191.8 [WRITE:191.8]
Total partitions : 1000000 [WRITE:1000000]
Total errors : 0 [WRITE:0]
total gc count : 5
total gc mb : 7351
total gc time (s) : 1
avg gc time(ms) : 147
stdev gc time(ms) : 34
Total operation time : 00:00:08

采用cql3协议写入

./cassandra-stress write n=1000000 cl=one -mode native cql3 -schema keyspace=”keyspace1” -log file=SINGLENODE5_write_1M_rows.log -node 10.21.21.10
测试结果:基本上和默认的模式是一样的,那么默认的模式是什么类型的?

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 123953 [WRITE:123953]
partition rate : 123953 [WRITE:123953]
row rate : 123953 [WRITE:123953]
latency mean : 1.6 [WRITE:1.6]
latency median : 1.4 [WRITE:1.4]
latency 95th percentile : 1.8 [WRITE:1.8]
latency 99th percentile : 2.5 [WRITE:2.5]
latency 99.9th percentile : 18.7 [WRITE:18.7]
latency max : 201.6 [WRITE:201.6]
Total partitions : 1000000 [WRITE:1000000]
Total errors : 0 [WRITE:0]
total gc count : 4
total gc mb : 5753
total gc time (s) : 1
avg gc time(ms) : 165
stdev gc time(ms) : 39
Total operation time : 00:00:08

可能是由于数量太小(100万还不小?可以设置为500万,差不多需要7分钟),更改内存,基本上看不出差别。

读取测试

读取20万条记录:./cassandra-stress read n=200000 -node 10.21.21.10 -log file=SINGLENODE2_read_2.log
测试结果:使用121个线程时,每秒最大可以读取8万条数据,平均每条读取时间1.4ms,99.9%的读取时间在2ms内,
读取时间最长的那条记录花了21ms,GC发生了一次,花费了14ms。
问题:为什么读取也会有GC? 因为读取时会先把数据放到内存,然后才会返回给客户端。

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 24220, 24220, 24220, 0.2, 0.1, 0.2, 0.2, 0.3, 40.3, 8.3, 0.01704, 0, 2, 69, 69, 5, 3380
8 threadCount, READ, 200000, 44524, 44524, 44524, 0.2, 0.2, 0.2, 0.2, 0.3, 26.7, 4.5, 0.01377, 0, 1, 25, 25, 0, 1640
16 threadCount, READ, 200000, 66583, 66583, 66583, 0.2, 0.2, 0.3, 0.3, 1.1, 32.3, 3.0, 0.01140, 0, 2, 16, 29, 2, 1646
24 threadCount, READ, 200000, 71464, 71464, 71464, 0.3, 0.3, 0.4, 0.5, 0.7, 19.1, 2.8, 0.02220, 0, 1, 13, 13, 0, 1638
36 threadCount, READ, 200000, 74215, 74215, 74215, 0.5, 0.5, 0.6, 0.8, 1.1, 16.2, 2.7, 0.02348, 0, 1, 15, 15, 0, 1637
54 threadCount, READ, 200000, 75220, 75220, 75220, 0.7, 0.7, 0.9, 1.0, 1.6, 16.4, 2.7, 0.00871, 0, 2, 28, 28, 0, 3276
81 threadCount, READ, 200000, 77820, 77820, 77820, 1.0, 1.0, 1.3, 1.6, 7.0, 17.6, 2.6, 0.02741, 0, 1, 15, 15, 0, 1638
121 threadCount, READ, 200000, 83228, 83228, 83228, 1.4, 1.4, 1.6, 1.7, 2.0, 21.0, 2.4, 0.00911, 0, 1, 14, 14, 0, 1638
181 threadCount, READ, 200000, 77173, 77173, 77173, 2.3, 2.4, 2.7, 3.2, 16.7, 18.6, 2.6, 0.01297, 0, 1, 14, 14, 0, 1638
271 threadCount, READ, 200000, 71700, 71700, 71700, 3.8, 3.6, 4.9, 5.7, 18.7, 20.6, 2.8, 0.02096, 0, 1, 13, 13, 0, 1638

上面是stress客户端和Cassandra运行在同一个节点,客户端本身消耗一定的资源,下面是客户端在集群之外的测试:
测试结果:线程数量在181时达到最大,因为客户端在集群里面本身开启一些测试线程,把客户端排除在外后,可用的线程数增大。
在181个线程下,每秒钟最大可以读取13万条,99.9%的读取时间在2.5ms,GC一次花了14ms。
可以看到跟上面的结果是差不多的,只不过因为客户端线程本身占用了一些线程,导致没有达到最大线程。

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 19522, 19522, 19522, 0.2, 0.2, 0.2, 0.3, 0.6, 37.4, 10.2, 0.01172, 0, 2, 67, 67, 3, 3265
8 threadCount, READ, 200000, 32361, 32361, 32361, 0.2, 0.2, 0.3, 0.3, 0.4, 26.4, 6.2, 0.00157, 0, 1, 25, 25, 0, 1638
16 threadCount, READ, 200000, 48620, 48620, 48620, 0.3, 0.3, 0.4, 0.5, 0.6, 18.9, 4.1, 0.01178, 0, 1, 17, 17, 0, 1636
24 threadCount, READ, 200000, 56026, 56026, 56026, 0.4, 0.4, 0.5, 0.6, 0.8, 17.7, 3.6, 0.00253, 0, 2, 32, 32, 1, 3276
36 threadCount, READ, 200000, 70923, 70923, 70923, 0.5, 0.5, 0.6, 0.7, 0.8, 16.9, 2.8, 0.00233, 0, 1, 15, 15, 0, 1638
54 threadCount, READ, 200000, 86152, 86152, 86152, 0.6, 0.6, 0.8, 0.8, 0.9, 16.8, 2.3, 0.01160, 0, 1, 15, 15, 0, 1638
81 threadCount, READ, 200000, 103282, 103282, 103282, 0.8, 0.7, 1.0, 1.0, 1.2, 21.5, 1.9, 0.00076, 0, 1, 15, 15, 0, 1638
121 threadCount, READ, 200000, 123607, 123607, 123607, 1.0, 0.9, 1.3, 1.5, 22.9, 31.2, 1.6, 0.01687, 0, 1, 14, 14, 0, 1638
181 threadCount, READ, 200000, 134298, 134298, 134298, 1.3, 1.2, 1.8, 2.1, 2.5, 27.6, 1.5, 0.00338, 0, 1, 14, 14, 0, 1638
271 threadCount, READ, 200000, 132761, 132761, 132761, 2.1, 1.8, 2.9, 3.2, 17.7, 25.8, 1.5, 0.01147, 0, 2, 28, 28, 0, 3276
406 threadCount, READ, 200000, 121970, 121970, 121970, 3.3, 3.1, 4.7, 5.6, 37.4, 39.2, 1.6, 0.04408, 0, 1, 14, 14, 0, 1638

问题:这里的线程数是什么含义?跟Cassandra本身的concurrent_reads有没有什么关系,如果更了concurrent_reads,这里的线程数会不会有影响?
答:这里的线程指的应该是客户端线程数,就好比多个客户端同时要查询Cassandra,实际中是API分布在多台机器上,每台机器都会查询Cassandra。所以线程数指的是可以允许多少个客户端同时查询。

又测试一次,看起来很上面差不多,不过不同的是.999的读取时间在81个线程-271个线程不断递增,而上面在181线程却降下来了。
实际上即使每次测试,结果也是不同的,不过可以测试多次,观察相同的现象,从而得出最优的线程。

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 20232, 20232, 20232, 0.2, 0.2, 0.2, 0.3, 0.3, 29.9, 9.9, 0.00875, 0, 2, 56, 56, 1, 3408
8 threadCount, READ, 200000, 31871, 31871, 31871, 0.2, 0.2, 0.3, 0.4, 0.4, 24.8, 6.3, 0.00596, 0, 1, 24, 24, 0, 1640
16 threadCount, READ, 200000, 48916, 48916, 48916, 0.3, 0.3, 0.4, 0.5, 0.5, 18.4, 4.1, 0.00362, 0, 1, 16, 16, 0, 1638
24 threadCount, READ, 200000, 57223, 57223, 57223, 0.4, 0.4, 0.5, 0.6, 0.6, 17.2, 3.5, 0.00869, 0, 1, 15, 15, 0, 1637
36 threadCount, READ, 200000, 72069, 72069, 72069, 0.5, 0.5, 0.6, 0.7, 0.7, 15.7, 2.8, 0.00479, 0, 1, 14, 14, 0, 1638
54 threadCount, READ, 200000, 84576, 84576, 84576, 0.6, 0.6, 0.8, 0.8, 8.8, 16.1, 2.4, 0.01512, 0, 2, 27, 27, 1, 3276
81 threadCount, READ, 200000, 97560, 97560, 97560, 0.8, 0.8, 1.0, 1.3, 10.5, 19.3, 2.0, 0.00667, 0, 1, 13, 13, 0, 1638
121 threadCount, READ, 200000, 124550, 124550, 124550, 1.0, 0.9, 1.3, 1.5, 11.3, 15.5, 1.6, 0.01291, 0, 1, 13, 13, 0, 1638
181 threadCount, READ, 200000, 128233, 128233, 128233, 1.4, 1.3, 1.9, 2.2, 15.2, 26.5, 1.6, 0.02911, 0, 1, 14, 14, 0, 1638
271 threadCount, READ, 200000, 129338, 129338, 129338, 2.1, 1.8, 3.0, 3.3, 36.4, 37.1, 1.5, 0.00817, 0, 1, 14, 14, 0, 1638
406 threadCount, READ, 200000, 122300, 122300, 122300, 3.3, 3.1, 4.4, 5.3, 30.9, 33.7, 1.6, 0.02184, 0, 1, 14, 14, 0, 1638

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 20314, 20314, 20314, 0.2, 0.2, 0.2, 0.3, 0.3, 217.3, 9.8, 0.00695, 0, 1, 205, 205, 0, 4514
8 threadCount, READ, 200000, 32903, 32903, 32903, 0.2, 0.2, 0.3, 0.3, 0.5, 11.1, 6.1, 0.05020, 0, 0, 0, 0, 0, 0
16 threadCount, READ, 200000, 50563, 50563, 50563, 0.3, 0.3, 0.4, 0.5, 0.5, 8.5, 4.0, 0.00313, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 200000, 57049, 57049, 57049, 0.4, 0.4, 0.5, 0.6, 0.8, 130.7, 3.5, 0.00710, 0, 1, 129, 129, 0, 4876
36 threadCount, READ, 200000, 70887, 70887, 70887, 0.5, 0.5, 0.6, 0.7, 1.2, 11.6, 2.8, 0.00415, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 200000, 86965, 86965, 86965, 0.6, 0.6, 0.8, 0.8, 1.5, 16.1, 2.3, 0.01355, 0, 0, 0, 0, 0, 0
81 threadCount, READ, 200000, 98004, 98004, 98004, 0.8, 0.8, 1.0, 1.1, 12.4, 97.3, 2.0, 0.01691, 0, 1, 95, 95, 0, 4896
121 threadCount, READ, 200000, 123605, 123605, 123605, 1.0, 0.9, 1.3, 1.5, 1.7, 32.3, 1.6, 0.02667, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 200000, 137328, 137328, 137328, 1.3, 1.2, 1.8, 2.1, 2.6, 13.4, 1.5, 0.00533, 0, 0, 0, 0, 0, 0
271 threadCount, READ, 200000, 116140, 116140, 116140, 2.3, 1.9, 3.1, 3.7, 100.8, 102.4, 1.7, 0.04059, 0, 1, 98, 98, 0, 4900
406 threadCount, READ, 200000, 117164, 117164, 117164, 3.5, 3.2, 4.6, 5.1, 82.8, 86.3, 1.7, 0.02663, 0, 0, 0, 0, 0, 0

没有预热读取

没有预热读取20万条数据:./cassandra-stress read n=200000 no-warmup -node 10.21.21.10 -log file=SINGLENODE3_read_2_nowarm.log
测试结果:以181个线程为例,每秒的读取数量差不了多少,但是没有预热情况下,99.9%的读取时间在41.5ms,不过仍有99%的读取时间在2.3ms。

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 19246, 19246, 19246, 0.2, 0.2, 0.3, 0.3, 0.8, 22.7, 10.4, 0.03294, 0, 1, 20, 20, 0, 1634
8 threadCount, READ, 200000, 31951, 31951, 31951, 0.2, 0.2, 0.3, 0.4, 0.4, 31.8, 6.3, 0.00502, 0, 1, 31, 31, 0, 1638
16 threadCount, READ, 199935, 49763, 49763, 49763, 0.3, 0.3, 0.4, 0.5, 0.5, 16.4, 4.0, 0.00809, 0, 1, 15, 15, 0, 1640
24 threadCount, READ, 200000, 57285, 57285, 57285, 0.4, 0.4, 0.5, 0.6, 0.7, 16.1, 3.5, 0.01034, 0, 1, 15, 15, 0, 1639
36 threadCount, READ, 200000, 71017, 71017, 71017, 0.5, 0.5, 0.6, 0.7, 0.8, 15.9, 2.8, 0.00060, 0, 2, 30, 30, 0, 3277
54 threadCount, READ, 200000, 84927, 84927, 84927, 0.6, 0.6, 0.8, 0.8, 1.0, 16.6, 2.4, 0.01203, 0, 1, 14, 14, 0, 1638
81 threadCount, READ, 200000, 102405, 102405, 102405, 0.8, 0.8, 1.0, 1.1, 1.4, 16.6, 2.0, 0.00724, 0, 1, 14, 14, 0, 1638
121 threadCount, READ, 200000, 113897, 113897, 113897, 1.1, 0.9, 1.3, 1.6, 52.6, 89.4, 1.8, 0.01850, 0, 1, 14, 14, 0, 1638
181 threadCount, READ, 200000, 121195, 121195, 121195, 1.5, 1.3, 1.9, 2.3, 41.5, 61.9, 1.6, 0.02605, 0, 1, 15, 15, 0, 1638
271 threadCount, READ, 200000, 131395, 131395, 131395, 2.1, 1.9, 2.9, 3.2, 17.5, 20.2, 1.5, 0.00003, 0, 1, 14, 14, 0, 1638
406 threadCount, READ, 200000, 120490, 120490, 120490, 3.4, 3.3, 4.4, 7.9, 24.9, 33.7, 1.7, 0.02581, 0, 2, 28, 28, 0, 3277
609 threadCount, READ, 200000, 113461, 113461, 113461, 5.4, 5.5, 6.9, 9.5, 41.3, 43.5, 1.8, 0.02642, 0, 1, 14, 14, 0, 1638

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 19459, 19459, 19459, 0.2, 0.2, 0.3, 0.3, 0.7, 13.6, 10.3, 0.03267, 0, 0, 0, 0, 0, 0
8 threadCount, READ, 200000, 31733, 31733, 31733, 0.2, 0.2, 0.3, 0.4, 0.4, 11.4, 6.3, 0.00728, 0, 0, 0, 0, 0, 0
16 threadCount, READ, 200000, 48927, 48927, 48927, 0.3, 0.3, 0.4, 0.5, 0.6, 9.0, 4.1, 0.00572, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 200000, 57267, 57267, 57267, 0.4, 0.4, 0.5, 0.6, 0.7, 10.2, 3.5, 0.00991, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 200000, 71712, 71712, 71712, 0.5, 0.5, 0.6, 0.7, 0.8, 38.8, 2.8, 0.00863, 0, 1, 37, 37, 0, 8782
54 threadCount, READ, 200000, 89064, 89064, 89064, 0.6, 0.6, 0.8, 0.8, 1.2, 13.5, 2.2, 0.00836, 0, 0, 0, 0, 0, 0
81 threadCount, READ, 200000, 101873, 101873, 101873, 0.8, 0.8, 1.0, 1.1, 1.6, 26.6, 2.0, 0.01203, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 200000, 128791, 128791, 128791, 0.9, 0.9, 1.3, 1.4, 4.5, 20.3, 1.6, 0.00524, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 200000, 136667, 136667, 136667, 1.3, 1.2, 1.8, 2.1, 5.3, 31.5, 1.5, 0.01082, 0, 0, 0, 0, 0, 0
271 threadCount, READ, 200000, 121611, 121611, 121611, 2.2, 1.9, 3.1, 3.8, 31.1, 32.0, 1.6, 0.01160, 0, 1, 21, 21, 0, 8783
406 threadCount, READ, 200000, 118972, 118972, 118972, 3.4, 3.6, 4.5, 5.5, 35.2, 36.3, 1.7, 0.00088, 0, 0, 0, 0, 0, 0

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 19715, 19715, 19715, 0.2, 0.2, 0.2, 0.3, 0.7, 14.3, 10.1, 0.02534, 0, 0, 0, 0, 0, 0
8 threadCount, READ, 200000, 32683, 32683, 32683, 0.2, 0.2, 0.3, 0.3, 0.4, 10.1, 6.1, 0.00510, 0, 0, 0, 0, 0, 0
16 threadCount, READ, 200000, 49059, 49059, 49059, 0.3, 0.3, 0.4, 0.5, 0.5, 9.0, 4.1, 0.00434, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 200000, 58071, 58071, 58071, 0.4, 0.4, 0.5, 0.6, 0.6, 8.2, 3.4, 0.01215, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 200000, 71800, 71800, 71800, 0.5, 0.5, 0.6, 0.7, 5.6, 10.1, 2.8, 0.00305, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 200000, 85405, 85405, 85405, 0.6, 0.6, 0.8, 0.8, 2.5, 15.0, 2.3, 0.01252, 0, 1, 12, 12, 0, 14734
81 threadCount, READ, 200000, 102288, 102288, 102288, 0.8, 0.8, 1.0, 1.1, 2.4, 10.3, 2.0, 0.00164, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 200000, 122329, 122329, 122329, 1.0, 0.9, 1.3, 1.5, 15.5, 92.1, 1.6, 0.00384, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 200000, 127409, 127409, 127409, 1.4, 1.2, 1.8, 2.0, 45.2, 83.2, 1.6, 0.02074, 0, 0, 0, 0, 0, 0
271 threadCount, READ, 200000, 129012, 129012, 129012, 2.1, 1.8, 3.0, 3.8, 24.2, 25.1, 1.5, 0.02108, 0, 0, 0, 0, 0, 0
406 threadCount, READ, 200000, 124542, 124542, 124542, 3.3, 3.4, 4.3, 5.1, 21.1, 22.3, 1.6, 0.01146, 0, 0, 0, 0, 0, 0
609 threadCount, READ, 200000, 114549, 114549, 114549, 5.4, 5.5, 6.9, 7.8, 26.4, 28.6, 1.7, 0.00217, 0, 0, 0, 0, 0, 0

G1_32G

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 18901, 18901, 18901, 0.2, 0.2, 0.3, 0.3, 0.7, 14.6, 10.6, 0.02816, 0, 0, 0, 0, 0, 0
8 threadCount, READ, 200000, 32102, 32102, 32102, 0.2, 0.2, 0.3, 0.3, 0.4, 13.4, 6.2, 0.00663, 0, 1, 13, 13, 0, 19629
16 threadCount, READ, 199859, 49739, 49739, 49739, 0.3, 0.3, 0.4, 0.4, 0.6, 6.3, 4.0, 0.00201, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 200000, 56403, 56403, 56403, 0.4, 0.4, 0.6, 0.6, 0.7, 8.5, 3.5, 0.01001, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 200000, 72080, 72080, 72080, 0.5, 0.5, 0.6, 0.6, 0.9, 9.7, 2.8, 0.00493, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 200000, 85258, 85258, 85258, 0.6, 0.6, 0.8, 0.9, 1.3, 20.8, 2.3, 0.00435, 0, 0, 0, 0, 0, 0
81 threadCount, READ, 200000, 96518, 96518, 96518, 0.8, 0.8, 1.0, 1.1, 1.2, 15.6, 2.1, 0.00669, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 200000, 120044, 120044, 120044, 1.0, 0.9, 1.3, 1.5, 14.7, 36.8, 1.7, 0.01983, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 200000, 129879, 129879, 129879, 1.4, 1.3, 1.7, 2.1, 23.6, 70.6, 1.5, 0.03941, 0, 0, 0, 0, 0, 0
271 threadCount, READ, 200000, 134088, 134088, 134088, 2.0, 1.8, 2.9, 3.3, 15.9, 16.6, 1.5, 0.00226, 0, 0, 0, 0, 0, 0
406 threadCount, READ, 200000, 120807, 120807, 120807, 3.4, 3.0, 4.6, 5.1, 51.4, 53.9, 1.7, 0.00992, 0, 1, 12, 12, 0, 19630
609 threadCount, READ, 200000, 101774, 101774, 101774, 6.1, 5.7, 7.0, 8.2, 227.4, 229.6, 2.0, 0.00759, 0, 0, 0, 0, 0, 0

给定一段时间测试最大可以读取多少条记录

测试3分钟能读取多少条记录:./cassandra-stress read duration=3m -node 10.21.21.10 -log file=SINGLENODE4_read_3min.log
测试结果:还是181个线程时达到最大值,总共读取了2500万,99.9%的读取时间为2.4ms,GC发生了144次,最长的GC时间花了2.6s。

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 3685598, 20476, 20476, 20476, 0.2, 0.2, 0.2, 0.3, 0.3, 57.0, 180.0, 0.00144, 0, 27, 757, 757, 11, 44103
8 threadCount, READ, 5828269, 32379, 32379, 32379, 0.2, 0.2, 0.3, 0.3, 0.4, 25.5, 180.0, 0.00112, 0, 36, 706, 706, 1, 58974
16 threadCount, READ, 8829104, 49050, 49050, 49050, 0.3, 0.3, 0.4, 0.5, 0.6, 21.7, 180.0, 0.00163, 0, 52, 979, 979, 1, 85196
24 threadCount, READ, 10249086, 56939, 56939, 56939, 0.4, 0.4, 0.5, 0.6, 0.7, 37.2, 180.0, 0.00170, 0, 61, 1140, 1152, 1, 98314
36 threadCount, READ, 12957013, 71983, 71983, 71983, 0.5, 0.5, 0.6, 0.6, 0.8, 21.1, 180.0, 0.00072, 0, 76, 1414, 1414, 1, 124523
54 threadCount, READ, 15659754, 86997, 86997, 86997, 0.6, 0.6, 0.7, 0.8, 1.0, 101.4, 180.0, 0.00075, 0, 91, 1691, 1691, 1, 149101
81 threadCount, READ, 18822803, 104569, 104569, 104569, 0.8, 0.7, 1.0, 1.0, 1.2, 36.3, 180.0, 0.00091, 0, 110, 2051, 2051, 1, 180233
121 threadCount, READ, 20570170, 114275, 114275, 114275, 1.0, 0.9, 1.2, 1.4, 1.6, 994.0, 180.0, 0.01009, 0, 118, 2185, 2185, 1, 193341
181 threadCount, READ, 24761012, 137557, 137557, 137557, 1.3, 1.2, 1.7, 1.9, 2.4, 150.7, 180.0, 0.00113, 0, 144, 2683, 2683, 1, 235942
271 threadCount, READ, 24495851, 136082, 136082, 136082, 2.0, 1.8, 2.8, 3.1, 11.6, 37.0, 180.0, 0.00120, 0, 141, 2642, 2642, 1, 231027
406 threadCount, READ, 20696670, 115470, 115470, 115470, 3.5, 3.1, 4.0, 4.5, 23.5, 883.7, 179.2, 0.01129, 0, 119, 2226, 2226, 1, 194980
609 threadCount, READ, 20807103, 115587, 115587, 115587, 5.3, 5.4, 6.4, 6.9, 26.1, 36.5, 180.0, 0.00097, 0, 121, 2266, 2266, 1, 198256

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 3584751, 19915, 19915, 19915, 0.2, 0.2, 0.2, 0.3, 0.3, 237.3, 180.0, 0.00161, 0, 12, 1953, 1953, 34, 103288
8 threadCount, READ, 5811977, 32289, 32289, 32289, 0.2, 0.2, 0.3, 0.3, 0.4, 121.8, 180.0, 0.00108, 0, 13, 1103, 1103, 13, 113422
16 threadCount, READ, 8852062, 49178, 49178, 49178, 0.3, 0.3, 0.4, 0.4, 0.5, 63.2, 180.0, 0.00176, 0, 17, 956, 956, 4, 149556
24 threadCount, READ, 10255674, 56976, 56976, 56976, 0.4, 0.4, 0.5, 0.6, 0.7, 58.2, 180.0, 0.00189, 0, 11, 541, 541, 1, 96735
36 threadCount, READ, 13003446, 72241, 72241, 72241, 0.5, 0.5, 0.6, 0.7, 0.8, 49.9, 180.0, 0.00082, 0, 13, 616, 616, 0, 114296
54 threadCount, READ, 15887097, 88261, 88261, 88261, 0.6, 0.6, 0.7, 0.8, 1.0, 107.7, 180.0, 0.00110, 0, 17, 794, 794, 0, 149463
81 threadCount, READ, 17262985, 95904, 95904, 95904, 0.8, 0.7, 1.0, 1.0, 1.2, 923.2, 180.0, 0.00918, 0, 18, 843, 843, 1, 158255
121 threadCount, READ, 21136848, 117427, 117427, 117427, 1.0, 0.9, 1.2, 1.4, 1.6, 977.7, 180.0, 0.01099, 0, 23, 1070, 1070, 1, 202215
181 threadCount, READ, 24958916, 138650, 138650, 138650, 1.3, 1.2, 1.6, 2.0, 2.3, 203.5, 180.0, 0.00150, 0, 26, 1198, 1198, 0, 228591
271 threadCount, READ, 21354485, 118630, 118630, 118630, 2.3, 1.8, 2.9, 3.2, 3.7, 911.6, 180.0, 0.01007, 0, 22, 1016, 1016, 0, 193423
406 threadCount, READ, 21890345, 121606, 121606, 121606, 3.3, 3.3, 4.5, 4.9, 5.9, 286.9, 180.0, 0.00107, 0, 23, 1060, 1060, 0, 202215
609 threadCount, READ, 20206343, 112250, 112250, 112250, 5.4, 5.4, 6.9, 7.3, 16.5, 77.0, 180.0, 0.00111, 0, 22, 1020, 1020, 0, 193374

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 3776268, 20979, 20979, 20979, 0.2, 0.2, 0.2, 0.2, 0.3, 62.2, 180.0, 0.00140, 0, 2, 121, 121, 1, 29133
8 threadCount, READ, 5840032, 32445, 32445, 32445, 0.2, 0.2, 0.3, 0.3, 0.4, 63.1, 180.0, 0.00112, 0, 3, 182, 182, 1, 43696
16 threadCount, READ, 8924201, 49579, 49579, 49579, 0.3, 0.3, 0.4, 0.5, 0.5, 63.3, 180.0, 0.00176, 0, 4, 246, 246, 1, 58256
24 threadCount, READ, 10387219, 57707, 57707, 57707, 0.4, 0.4, 0.5, 0.6, 0.7, 87.3, 180.0, 0.00197, 0, 5, 297, 297, 20, 72872
36 threadCount, READ, 12901814, 71676, 71676, 71676, 0.5, 0.5, 0.6, 0.6, 0.7, 19.5, 180.0, 0.00079, 0, 6, 91, 91, 1, 88381
54 threadCount, READ, 15457962, 85876, 85876, 85876, 0.6, 0.6, 0.8, 0.8, 0.9, 142.2, 180.0, 0.00079, 0, 7, 82, 82, 1, 103152
81 threadCount, READ, 16689834, 92720, 92720, 92720, 0.9, 0.8, 1.0, 1.1, 1.2, 1028.4, 180.0, 0.01156, 0, 8, 87, 87, 0, 117888
121 threadCount, READ, 23132477, 128511, 128511, 128511, 0.9, 0.9, 1.3, 1.4, 1.7, 20.0, 180.0, 0.00100, 0, 11, 121, 121, 0, 162096
181 threadCount, READ, 24764144, 137574, 137574, 137574, 1.3, 1.2, 1.7, 2.0, 2.3, 24.9, 180.0, 0.00079, 0, 11, 121, 121, 0, 162096
271 threadCount, READ, 21903620, 121680, 121680, 121680, 2.2, 1.8, 2.8, 3.1, 3.6, 929.1, 180.0, 0.01170, 0, 10, 111, 111, 0, 147360
406 threadCount, READ, 20649352, 114685, 114685, 114685, 3.5, 2.9, 4.3, 4.7, 5.5, 923.6, 180.1, 0.01094, 0, 10, 112, 112, 0, 147360

G1_32G

1
2
3
4
5
6
7
8
9
10
11
12
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 3612926, 20072, 20072, 20072, 0.2, 0.2, 0.2, 0.3, 0.3, 67.7, 180.0, 0.00157, 0, 2, 132, 132, 0, 38800
8 threadCount, READ, 5744438, 31914, 31914, 31914, 0.2, 0.2, 0.3, 0.3, 0.3, 67.9, 180.0, 0.00106, 0, 4, 263, 263, 1, 77536
16 threadCount, READ, 8824045, 49022, 49022, 49022, 0.3, 0.3, 0.4, 0.4, 0.5, 69.4, 180.0, 0.00145, 0, 4, 264, 264, 1, 77492
24 threadCount, READ, 9937324, 55207, 55207, 55207, 0.4, 0.4, 0.6, 0.6, 0.7, 94.4, 180.0, 0.00163, 0, 6, 284, 284, 28, 116877
36 threadCount, READ, 13117518, 72873, 72873, 72873, 0.5, 0.5, 0.6, 0.6, 0.7, 20.0, 180.0, 0.00073, 0, 6, 79, 79, 2, 117790
54 threadCount, READ, 15457163, 85872, 85872, 85872, 0.6, 0.6, 0.8, 0.8, 0.9, 128.8, 180.0, 0.00117, 0, 8, 85, 85, 1, 157056
81 threadCount, READ, 16216625, 90090, 90090, 90090, 0.9, 0.8, 1.0, 1.1, 1.2, 1069.8, 180.0, 0.00725, 0, 9, 94, 94, 0, 176688
121 threadCount, READ, 22825729, 126807, 126807, 126807, 0.9, 0.9, 1.3, 1.4, 1.6, 959.0, 180.0, 0.00100, 0, 12, 129, 129, 0, 235584
181 threadCount, READ, 24977336, 138758, 138758, 138758, 1.3, 1.2, 1.6, 1.9, 2.3, 29.0, 180.0, 0.00089, 0, 12, 128, 128, 0, 235584
271 threadCount, READ, 22096073, 122749, 122749, 122749, 2.2, 1.8, 2.8, 3.2, 3.7, 839.5, 180.0, 0.01227, 0, 12, 129, 129, 0, 235584
406 threadCount, READ, 22520223, 125103, 125103, 125103, 3.2, 3.1, 4.2, 4.7, 5.3, 136.5, 180.0, 0.00141, 0, 11, 119, 119, 0, 215952

混合模式读大于写

./cassandra-stress mixed ratio(write=1,read=3) n=100000 cl=ONE -pop dist=UNIFORM(1..1000000) \
-schema keyspace=”keyspace1” -mode native cql3 -rate threads>=16 threads\<=256 -log file=SINGLENODE6_mixed_autorate_50r50w_1M.log -node 10.21.21.10

population=10万,其中读和写的比例是3:1,比如读7万条,写2万条,这种是读比较多的场景。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
16 threadCount, READ, 75251, 34505, 34505, 34505, 0.3, 0.3, 0.4, 0.5, 0.7, 46.8, 2.2, 0.03844, 0, 1, 45, 45, 0, 1712
16 threadCount, WRITE, 24749, 11348, 11348, 11348, 0.3, 0.3, 0.4, 0.5, 0.8, 46.7, 2.2, 0.03844, 0, 1, 45, 45, 0, 1712
16 threadCount, total, 100000, 45853, 45853, 45853, 0.3, 0.3, 0.4, 0.5, 0.8, 46.8, 2.2, 0.03844, 0, 1, 45, 45, 0, 1712
24 threadCount, READ, 74844, 40916, 40916, 40916, 0.4, 0.4, 0.5, 0.6, 0.8, 62.0, 1.8, 0.03023, 0, 1, 60, 60, 0, 1602
24 threadCount, WRITE, 25156, 13754, 13754, 13754, 0.4, 0.4, 0.5, 0.6, 0.9, 62.0, 1.8, 0.03023, 0, 1, 60, 60, 0, 1602
24 threadCount, total, 100000, 54668, 54668, 54668, 0.4, 0.4, 0.5, 0.6, 0.7, 62.0, 1.8, 0.03023, 0, 1, 60, 60, 0, 1602
36 threadCount, READ, 74519, 54694, 54694, 54694, 0.5, 0.5, 0.6, 0.7, 1.6, 3.6, 1.4, 0.00051, 0, 0, 0, 0, 0, 0
36 threadCount, WRITE, 25481, 18702, 18702, 18702, 0.5, 0.5, 0.6, 0.7, 1.7, 3.6, 1.4, 0.00051, 0, 0, 0, 0, 0, 0
36 threadCount, total, 100000, 73396, 73396, 73396, 0.5, 0.5, 0.6, 0.7, 1.5, 3.6, 1.4, 0.00051, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 75429, 62268, 62268, 62268, 0.6, 0.6, 0.8, 1.0, 3.8, 51.4, 1.2, 0.00555, 0, 1, 49, 49, 0, 1621
54 threadCount, WRITE, 24571, 20284, 20284, 20284, 0.6, 0.6, 0.8, 1.0, 25.6, 51.3, 1.2, 0.00555, 0, 1, 49, 49, 0, 1621
54 threadCount, total, 100000, 82551, 82551, 82551, 0.6, 0.6, 0.8, 1.0, 7.2, 51.4, 1.2, 0.00555, 0, 1, 49, 49, 0, 1621
81 threadCount, READ, 74648, 77082, 77082, 77082, 0.8, 0.7, 1.0, 1.2, 2.5, 51.3, 1.0, 0.00000, 0, 1, 48, 48, 0, 1606
81 threadCount, WRITE, 25352, 26181, 26181, 26181, 0.8, 0.7, 1.0, 1.2, 3.4, 51.2, 1.0, 0.00000, 0, 1, 48, 48, 0, 1606
81 threadCount, total, 100000, 103261, 103261, 103261, 0.8, 0.7, 1.0, 1.2, 2.4, 51.3, 1.0, 0.00000, 0, 1, 48, 48, 0, 1606
121 threadCount, READ, 74735, 85846, 85846, 85846, 1.0, 0.9, 1.1, 1.3, 2.5, 54.0, 0.9, 0.00000, 0, 1, 51, 51, 0, 1609
121 threadCount, WRITE, 25265, 29358, 29358, 29358, 1.0, 0.8, 1.2, 1.4, 53.6, 53.9, 0.9, 0.00000, 0, 1, 51, 51, 0, 1609
121 threadCount, total, 100000, 114867, 114867, 114867, 1.0, 0.9, 1.1, 1.3, 2.4, 54.0, 0.9, 0.00000, 0, 1, 51, 51, 0, 1609
181 threadCount, READ, 74534, 104771, 104771, 104771, 1.3, 1.3, 1.7, 1.9, 5.3, 7.7, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, WRITE, 25466, 35798, 35798, 35798, 1.3, 1.2, 1.7, 2.0, 4.9, 8.3, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, total, 100000, 140568, 140568, 140568, 1.3, 1.2, 1.7, 1.9, 5.1, 8.3, 0.7, 0.00000, 0, 0, 0, 0, 0, 0

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
16 threadCount, READ, 75381, 34242, 34242, 34242, 0.3, 0.3, 0.4, 0.5, 0.7, 18.8, 2.2, 0.00397, 0, 0, 0, 0, 0, 0
16 threadCount, WRITE, 24619, 11183, 11183, 11183, 0.3, 0.3, 0.4, 0.5, 0.8, 18.8, 2.2, 0.00397, 0, 0, 0, 0, 0, 0
16 threadCount, total, 100000, 45425, 45425, 45425, 0.3, 0.3, 0.4, 0.5, 0.7, 18.8, 2.2, 0.00397, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 74898, 42757, 42757, 42757, 0.4, 0.4, 0.5, 0.6, 0.8, 17.6, 1.8, 0.02794, 0, 0, 0, 0, 0, 0
24 threadCount, WRITE, 25102, 14331, 14331, 14331, 0.4, 0.4, 0.5, 0.6, 0.9, 17.6, 1.8, 0.02794, 0, 0, 0, 0, 0, 0
24 threadCount, total, 100000, 57087, 57087, 57087, 0.4, 0.4, 0.5, 0.6, 0.8, 17.6, 1.8, 0.02794, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 75189, 54221, 54221, 54221, 0.5, 0.5, 0.6, 0.7, 1.7, 8.3, 1.4, 0.00024, 0, 0, 0, 0, 0, 0
36 threadCount, WRITE, 24811, 17892, 17892, 17892, 0.5, 0.5, 0.6, 0.7, 1.7, 5.2, 1.4, 0.00024, 0, 0, 0, 0, 0, 0
36 threadCount, total, 100000, 72113, 72113, 72113, 0.5, 0.5, 0.6, 0.7, 1.7, 8.3, 1.4, 0.00024, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 74941, 63319, 63319, 63319, 0.6, 0.6, 0.8, 1.0, 1.5, 24.7, 1.2, 0.02843, 0, 0, 0, 0, 0, 0
54 threadCount, WRITE, 25059, 21175, 21175, 21175, 0.6, 0.6, 0.8, 1.0, 2.0, 24.5, 1.2, 0.02843, 0, 0, 0, 0, 0, 0
54 threadCount, total, 100000, 84492, 84492, 84492, 0.6, 0.6, 0.8, 1.0, 1.4, 24.7, 1.2, 0.02843, 0, 0, 0, 0, 0, 0
81 threadCount, READ, 75009, 80336, 80336, 80336, 0.7, 0.7, 1.0, 1.0, 1.1, 7.2, 0.9, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, WRITE, 24991, 26766, 26766, 26766, 0.7, 0.7, 1.0, 1.1, 1.3, 3.1, 0.9, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, total, 100000, 107102, 107102, 107102, 0.7, 0.7, 1.0, 1.0, 1.1, 7.2, 0.9, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 75063, 91080, 91080, 91080, 1.0, 0.9, 1.3, 1.5, 3.2, 29.6, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, WRITE, 24937, 30259, 30259, 30259, 1.0, 0.9, 1.3, 1.7, 28.2, 29.6, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, total, 100000, 121338, 121338, 121338, 1.0, 0.9, 1.3, 1.5, 13.5, 29.6, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 75342, 104114, 104114, 104114, 1.3, 1.3, 1.7, 1.9, 3.9, 7.6, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, WRITE, 24658, 34075, 34075, 34075, 1.3, 1.3, 1.7, 2.0, 4.9, 8.0, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, total, 100000, 138189, 138189, 138189, 1.3, 1.3, 1.7, 1.9, 3.3, 8.0, 0.7, 0.00000, 0, 0, 0, 0, 0, 0

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
 16 threadCount, READ,          74594,   33655,   33655,   33655,     0.3,     0.3,     0.5,     0.5,     0.8,    23.1,    2.2,  0.00911,      0,      0,       0,       0,       0,       0
16 threadCount, WRITE, 25406, 11467, 11467, 11467, 0.3, 0.3, 0.5, 0.5, 0.8, 9.9, 2.2, 0.00911, 0, 0, 0, 0, 0, 0
16 threadCount, total, 100000, 45117, 45117, 45117, 0.3, 0.3, 0.5, 0.5, 0.8, 23.1, 2.2, 0.00911, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 74649, 40798, 40798, 40798, 0.4, 0.4, 0.5, 0.6, 0.7, 13.7, 1.8, 0.01581, 0, 0, 0, 0, 0, 0
24 threadCount, WRITE, 25351, 13861, 13861, 13861, 0.4, 0.4, 0.5, 0.6, 0.7, 13.7, 1.8, 0.01581, 0, 0, 0, 0, 0, 0
24 threadCount, total, 100000, 54653, 54653, 54653, 0.4, 0.4, 0.5, 0.6, 0.7, 13.7, 1.8, 0.01581, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 75378, 53898, 53898, 53898, 0.5, 0.5, 0.6, 0.7, 1.6, 3.2, 1.4, 0.01693, 0, 0, 0, 0, 0, 0
36 threadCount, WRITE, 24622, 17606, 17606, 17606, 0.5, 0.5, 0.6, 0.7, 1.4, 2.9, 1.4, 0.01693, 0, 0, 0, 0, 0, 0
36 threadCount, total, 100000, 71503, 71503, 71503, 0.5, 0.5, 0.6, 0.7, 1.4, 3.2, 1.4, 0.01693, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 75012, 61606, 61606, 61606, 0.6, 0.6, 0.8, 1.0, 3.1, 30.7, 1.2, 0.02550, 0, 0, 0, 0, 0, 0
54 threadCount, WRITE, 24988, 20522, 20522, 20522, 0.6, 0.6, 0.8, 1.0, 3.8, 30.5, 1.2, 0.02550, 0, 0, 0, 0, 0, 0
54 threadCount, total, 100000, 82128, 82128, 82128, 0.6, 0.6, 0.8, 1.0, 3.4, 30.7, 1.2, 0.02550, 0, 0, 0, 0, 0, 0
81 threadCount, READ, 75469, 76118, 76118, 76118, 0.8, 0.8, 1.0, 1.1, 1.3, 3.0, 1.0, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, WRITE, 24531, 24748, 24748, 24748, 0.8, 0.8, 1.0, 1.1, 1.6, 2.6, 1.0, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, total, 100000, 100860, 100860, 100860, 0.8, 0.8, 1.0, 1.1, 1.3, 3.0, 1.0, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 75006, 90799, 90799, 90799, 1.0, 0.9, 1.3, 1.5, 31.1, 31.5, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, WRITE, 24994, 30257, 30257, 30257, 1.0, 0.9, 1.3, 1.5, 31.3, 31.5, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, total, 100000, 121055, 121055, 121055, 1.0, 0.9, 1.3, 1.5, 5.5, 31.5, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 75454, 103545, 103545, 103545, 1.3, 1.3, 1.7, 2.0, 2.2, 4.5, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, WRITE, 24546, 33685, 33685, 33685, 1.3, 1.3, 1.8, 2.1, 3.0, 4.5, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, total, 100000, 137229, 137229, 137229, 1.3, 1.3, 1.7, 2.0, 2.2, 4.5, 0.7, 0.00000, 0, 0, 0, 0, 0, 0

混合模式写大于读

./cassandra-stress mixed ratio(write=3,read=1) n=1000000 cl=ONE -pop dist=UNIFORM(1..10000000) -schema keyspace=”keyspace1” -mode native cql3 -rate threads>=16 threads\<=256 -log file=SINGLENODE6_mixed_write3read1_1M.log -node 10.21.21.10

自定义schema

http://www.datastax.com/dev/blog/improved-cassandra-2-1-stress-tool-benchmark-any-schema

自定义的YAML配置文件包括四部分:

  1. DDL:表结构
  2. 列的分布:定义每个列的大小,和每个Partition有多少列等
  3. 写入数据的分布:数据写入
  4. DML:数据查询

模拟数据一般考虑生成多少行(总的数据量和每行的数据量),每一列的数据大小和数据分布。

列分布是为了模拟实际的数据:model the size of the data in the column, the number of unique values, and the clustering of them withing a given partition.
分别对应了columnspec下每个column的size,population,cluster。当然并不是说每个列都要定义这三个选项,只有排序键cluster key才需要定义cluster(比如下文的publish_date)。

  • Size distribution – Defines the distribution of sizes for text, blob, set and list types (default of UNIFORM(4..8))
  • Population distribution – Defines the distribution of unique values for the column values (default of UNIFORM(1..100B))
  • Cluster distribution – Defines the distribution for the number of clustering prefixes within a given partition (default of FIXED(1))

B是Billion十亿,默认值100B=1000亿,M是Million百万。UNIFORM:均匀分布/平均分布,Gaussion:高斯分布/正太分布,中间高两边低
默认的size,population,cluster表示:长度=[4,8]均匀分布,数据量=[1,1000万]均匀分布,排序键=1,表示每一个主键只有一行数据。

size表示每一列的长度(比如默认每一列长度在4-8之间),population表示不同列的数量,比如n=10万,population=1000,表示在产生10万条数据,但是不同的数据只有1000条,也就是每条数据实际上有多个列。不过为什么UNIFORM(1..100B)是个范围而不是精确的数字呢?比如n=10万,而UNIFORM(1..100B),其中最大值100B=1000亿,那么数据是怎么分布的呢?

插入分布:插入数据要指定row-key,对应Cassandra的Partition这个概念,每一行row-key都对应了一个Partition。 批量插入数据的方式有LOGGED和UNLOGGED两种类型。

  • Partition distribution:The number of partitions to update per batch (default FIXED(1))
  • select distribution ratio:The ratio of rows each partition should insert as a proportion of the total possible rows for the partition (as defined by the clustering distribution columns). default FIXED(1)/1

列分布决定数据的分布,插入分布决定插入的方式。可以不需要配置插入分布,插入数据时会自动符合列分布。

示例

说明:在tools/bin下创建blogpost1.yaml配置文件。如果表结果已经存在,只需要定义keyspace和table就可以,
不需要定义keyspace_definition和table_definition。由于目前还是单节点测试,所以副本数=1。

表结构说明: blogposts存储博客,主键是domain,排序键是发布时间。在实际中domain是实际的站点,url是博客的链接地址。
比如domain=http://blog.csdn.net,在同一个站点发布博客的所有作者都放在同一行。这种一般用于现在非常流畅的聚合类应用。

目标:define a model for a simple app to hold blog posts for multiple websites 多个站点的帖子

而如果对于某个博客站点自己做的应用他不需要聚合其他站点的博客,就可以把domain理解成博客作者的主页。
比如我在CSDN博客于2015-01-27发表了一篇关于Hello Flink的入门文章:

domain:http://blog.csdn.net/zqhxuyuan
publish-date:2015-01-27
url: http://blog.csdn.net/zqhxuyuan/article/details/43196423

又在2016-05-18发表了一篇Apache GraphX的图文详解文章:

domain:http://blog.csdn.net/zqhxuyuan
publish-date:2016-05-18
url:http://blog.csdn.net/zqhxuyuan/article/details/48730515

在Cassandra相同key的不同记录以列式存储,实际存储结果类似:

row-key 2015-01-27:url:author:title:body 2016-05-18:url:author:title:body
http://blog.csdn.net/zqhxuyuan http://blog.csdn.net/zqhxuyuan/article/details/43196423:zqhxuyuan:Hello Flink:… http://blog.csdn.net/zqhxuyuan/article/details/48730515:zqhxuyuan:Apache GraphX 图文详解:…

集群环境下,可以修改副本数=3,不过可以先测试1个副本,然后再测试3个副本

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
### DDL ###
###########

# Keyspace Name
keyspace: stresscql

# The CQL for creating a keyspace (optional if it already exists)
keyspace_definition: |
CREATE KEYSPACE stresscql WITH replication = {'class': 'SimpleStrategy', 'replication_factor': 1};
# Table name
table: blogposts

# The CQL for creating a table you wish to stress (optional if it already exists)
table_definition: |
CREATE TABLE blogposts (
domain text,
published_date timeuuid,
url text,
author text,
title text,
body text,
PRIMARY KEY(domain, published_date)
) WITH CLUSTERING ORDER BY (published_date DESC)
AND compaction = { 'class':'LeveledCompactionStrategy' }
AND comment='A table to hold blog posts'

### Column Distribution Specifications ###
##########################################

columnspec:
- name: domain
size: gaussian(5..100) #domain names are relatively short
population: uniform(1..10M) #10M possible domains to pick from

- name: published_date #每个站点最多允许1000个帖子,用排序键决定列的数量
cluster: fixed(1000) #under each domain we will have max 1000 posts

- name: url
size: uniform(30..300)

- name: title #titles shouldn't go beyond 200 chars
size: gaussian(10..200)

- name: author
size: uniform(5..20) #author names should be short

- name: body
size: gaussian(100..5000) #the body of the blog post can be long

### Batch Ratio Distribution Specifications ###
###############################################

insert: #每个batch只插入一次(?), 比如可以用insert=1,或者insert=2,...
partitions: fixed(1) #Our partition key is the domain so only insert one per batch
select: fixed(1)/1000 #We have 1000 posts per domain so 1/1000 will allow 1 post per batch
batchtype: UNLOGGED #Unlogged batches


### A list of queries you wish to run against the schema ###
############################################################

queries:
singlepost:
cql: select * from blogposts where domain = ? LIMIT 1
fields: samerow
timeline:
cql: select url, title, published_date from blogposts where domain = ? LIMIT 10
fields: samerow

数据分布测试:

./cassandra-stress user profile=blogpost.yaml ops(insert=1) -rate threads=100 -node 10.21.21.130

1
2
3
4
5
6
7
8
9
10
  domain text,
published_date timeuuid,

columspec:
- name: domain
size: gaussian(5..20)
population: uniform(1..100)

- name: published_date
cluster: fixed(10)

数据特征:总数=1000条,相同domain不同published_date有10条domain一共有100条,所以总数=100*10=1000
字段的长度和类型也有关系,比如published_date是uuid类型,长度是固定的,即使指定size也不起作用。
而domain和autor为文本类型,支持size的分布方式,比如domain是高斯分布,5-20中间的部分12左右最多。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
[admin@cass021133 bin]$ cqlsh 10.21.21.130 -e "select domain,published_date,author from stresscql.blogposts1;"
...
;"{(\x06\x17}N\x01Y-ic:R | 29124890-da3b-11d9-4eb1-f78328592aad | |\x15\x0b\x17\x15-\x1cE\x1bYY\x1a
;"{(\x06\x17}N\x01Y-ic:R | 70758610-c81c-11d9-4eb1-f78328592aad | n4Q\x05\x02P\x1e0
;"{(\x06\x17}N\x01Y-ic:R | ce7cac90-9d72-11d9-4eb1-f78328592aad | \x07\x18\x0ew[
;"{(\x06\x17}N\x01Y-ic:R | 262528f0-ddf2-11d3-4eb1-f78328592aad | .x[ChI+}
;"{(\x06\x17}N\x01Y-ic:R | c5e60070-55de-11d0-4eb1-f78328592aad | p;.\x12\nKn[N-\x17^tc
;"{(\x06\x17}N\x01Y-ic:R | db0601c0-cb63-11c2-4eb1-f78328592aad | z$y\x06$v+DwMqsA\x00=\x08/3\x12\x1b
;"{(\x06\x17}N\x01Y-ic:R | 7b715db0-857b-11bc-4eb1-f78328592aad | \x1138\x15\x1e2\x13`\x19/@9M\x00\x088;
;"{(\x06\x17}N\x01Y-ic:R | ac53f6d0-254b-11bc-4eb1-f78328592aad | )\x0f\x14MWKc7t\x01qv]JZ[\x19~
;"{(\x06\x17}N\x01Y-ic:R | 468457e0-829e-11b7-4eb1-f78328592aad | 7\x02%\tn=y\x00\x19/M\x15\x1e"E
;"{(\x06\x17}N\x01Y-ic:R | 20466680-2370-11b4-4eb1-f78328592aad | >_u}x\x1cSFu\x0e)\x1ey\x10
(1000 rows)

[admin@cass021133 bin]$ cqlsh 10.21.21.130 -e "select distinct domain from stresscql.blogposts1;"
\x05!GRpp\x1dD\x1fuUpS
srP}.6R\x1c"xd)t
Mw}o^\x16zFQ -!y
;"{(\x06\x17}N\x01Y-ic:R
(100 rows)

只有写入

./cassandra-stress user profile=./blogpost1.yaml ops(insert=1) -node 10.21.21.10 -log file=SINGLENODE7_blog_insert.log

默认没有其他选项,会从4个线程开始,并逐步增加线程数量,直到达到一个限制。默认的插入模式协议采用native transport,以及使用PreparedStatement。
这里跟标准测试不同的是,没有指定n=?,因为在profile YAML配置文件中,insert部分已经可以说明数据将如何被插入。

参考文章中作者的笔记本单台机器,可以达到401个线程,每秒8500。下面的测试结果中线程的上限只有81个,不过每秒可以写入2.5万条记录。

1
2
3
4
5
6
7
8
            id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 244201, 8098, 8098, 11068, 0.4, 0.4, 0.6, 0.7, 1.0, 380.9, 30.2, 0.00348, 0, 4, 1207, 1207, 53, 5233
8 threadCount, insert, 429876, 14162, 14162, 19369, 0.5, 0.4, 0.7, 0.8, 1.4, 265.7, 30.4, 0.01082, 0, 7, 1606, 1606, 20, 9419
16 threadCount, insert, 658771, 21750, 21750, 29780, 0.6, 0.5, 0.8, 1.0, 2.2, 279.1, 30.3, 0.00910, 0, 12, 2182, 2266, 59, 16897
24 threadCount, insert, 754842, 24767, 24767, 33897, 0.8, 0.7, 1.0, 1.5, 4.5, 232.1, 30.5, 0.01926, 0, 13, 2256, 2256, 46, 19254
36 threadCount, insert, 768148, 25357, 25357, 34689, 1.3, 1.1, 1.5, 2.0, 3.9, 238.5, 30.3, 0.01159, 0, 13, 2095, 2185, 39, 19013
54 threadCount, insert, 761943, 25102, 25102, 34326, 2.0, 1.7, 2.1, 2.7, 25.4, 263.5, 30.4, 0.01724, 0, 13, 2076, 2152, 38, 19154
81 threadCount, insert, 755494, 24711, 24711, 33815, 3.2, 2.7, 3.1, 4.1, 148.3, 270.2, 30.6, 0.01210, 0, 13, 2227, 2227, 48, 19182
1
2
3
4
5
6
7
8
9
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 251178, 8245, 8245, 11278, 0.4, 0.4, 0.6, 0.7, 1.0, 274.2, 30.5, 0.00551, 0, 4, 902, 902, 43, 5347
8 threadCount, insert, 427090, 13995, 13995, 19119, 0.5, 0.4, 0.7, 0.8, 1.6, 235.4, 30.5, 0.00936, 0, 9, 1557, 1557, 46, 13130
16 threadCount, insert, 677353, 22261, 22261, 30444, 0.6, 0.5, 0.8, 1.1, 2.2, 239.3, 30.4, 0.00837, 0, 11, 2080, 2080, 40, 14631
24 threadCount, insert, 762573, 24999, 24999, 34192, 0.8, 0.7, 1.0, 1.2, 2.0, 246.0, 30.5, 0.00974, 0, 12, 2131, 2131, 38, 15912
36 threadCount, insert, 772075, 25215, 25215, 34509, 1.3, 1.1, 1.5, 2.3, 6.4, 250.1, 30.6, 0.00550, 0, 12, 2149, 2149, 31, 15900
54 threadCount, insert, 780249, 25669, 25669, 35082, 2.0, 1.7, 2.1, 2.6, 22.5, 279.1, 30.4, 0.00497, 0, 12, 2358, 2358, 24, 15793
81 threadCount, insert, 781505, 25519, 25519, 34884, 3.0, 2.7, 3.1, 4.1, 173.6, 310.7, 30.6, 0.01380, 0, 13, 2137, 2137, 37, 18920
121 threadCount, insert, 770318, 25009, 25009, 34189, 4.7, 4.0, 4.5, 5.4, 185.7, 277.4, 30.8, 0.01570, 0, 12, 2152, 2152, 36, 15916

G1_16G

1
2
3
4
5
6
7
8
9
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 248881, 8178, 8178, 11171, 0.4, 0.4, 0.6, 0.7, 1.1, 373.4, 30.4, 0.01796, 0, 2, 704, 704, 11, 5843
8 threadCount, insert, 446173, 14559, 14559, 19911, 0.5, 0.4, 0.7, 0.8, 1.2, 324.8, 30.6, 0.03065, 0, 2, 532, 532, 57, 7486
16 threadCount, insert, 690320, 22697, 22697, 31027, 0.6, 0.5, 0.8, 1.0, 2.1, 335.4, 30.4, 0.00967, 0, 4, 937, 937, 74, 14148
24 threadCount, insert, 767142, 25009, 25009, 34198, 0.8, 0.7, 1.0, 1.3, 2.9, 450.4, 30.7, 0.01082, 0, 4, 1506, 1506, 68, 18804
36 threadCount, insert, 779673, 25765, 25765, 35240, 1.3, 1.1, 1.5, 2.5, 16.4, 284.5, 30.3, 0.01161, 0, 5, 1062, 1062, 43, 14792
54 threadCount, insert, 806073, 25927, 25927, 35449, 2.0, 1.7, 2.2, 2.8, 8.2, 466.0, 31.1, 0.00649, 0, 5, 1667, 1667, 87, 18160
81 threadCount, insert, 804974, 26236, 26236, 35850, 3.0, 2.7, 3.1, 3.9, 44.3, 539.6, 30.7, 0.00619, 0, 4, 1155, 1155, 155, 15884
121 threadCount, insert, 767591, 25302, 25302, 34605, 4.7, 4.0, 4.6, 5.4, 111.4, 394.4, 30.3, 0.01340, 0, 6, 1585, 1585, 78, 16745

G1_24G

1
2
3
4
5
6
7
8
9
10
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 257051, 8524, 8524, 11675, 0.4, 0.3, 0.6, 0.7, 0.9, 224.7, 30.2, 0.00582, 0, 0, 0, 0, 0, 0
8 threadCount, insert, 450270, 14830, 14830, 20289, 0.4, 0.4, 0.7, 0.8, 1.4, 330.0, 30.4, 0.00674, 0, 1, 327, 327, 0, 13398
16 threadCount, insert, 702064, 23078, 23078, 31593, 0.6, 0.5, 0.8, 1.0, 2.0, 401.3, 30.4, 0.00763, 0, 2, 658, 658, 69, 26291
24 threadCount, insert, 798352, 26369, 26369, 36088, 0.8, 0.7, 1.0, 1.4, 2.3, 250.0, 30.3, 0.00953, 0, 1, 231, 231, 0, 12944
36 threadCount, insert, 824159, 27317, 27317, 37329, 1.2, 1.1, 1.5, 1.9, 3.5, 242.5, 30.2, 0.00507, 0, 2, 248, 248, 44, 27657
54 threadCount, insert, 812952, 26948, 26948, 36838, 1.9, 1.7, 2.1, 2.8, 8.7, 235.6, 30.2, 0.00507, 0, 1, 206, 206, 0, 13159
81 threadCount, insert, 826216, 27214, 27214, 37200, 2.9, 2.7, 3.1, 4.2, 25.3, 238.6, 30.4, 0.00570, 0, 2, 232, 232, 44, 27827
121 threadCount, insert, 834614, 27495, 27495, 37594, 4.3, 4.0, 4.5, 5.0, 23.9, 263.9, 30.4, 0.00342, 0, 2, 456, 456, 25, 26020
181 threadCount, insert, 795024, 25577, 25577, 34996, 6.9, 6.1, 6.7, 7.7, 113.7, 444.3, 31.1, 0.01847, 0, 1, 429, 429, 0, 11326

读写混合模式

类似于mixed模式指定读和写的线程数,自定义的ops包括insert和多种查询方式,比如blogpost1.yaml的queries定义了两种查询模式。

./cassandra-stress user profile=./blogpost1.yaml ops(singlepost=2,timeline=1,insert=1) -node 10.21.21.10 -log file=SINGLENODE8_blog_mixed.log

singlepost的查询是timeline查询和insert写入的两倍,比如查询singlepost次数=71309,写入insert=34965,查询timeline=35761。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 34965, 1118, 1118, 1528, 0.4, 0.4, 0.7, 0.8, 1.5, 56.7, 31.3, 0.00905, 0, 8, 413, 413, 32, 12700
4 threadCount, singlepost, 71309, 2280, 2166, 2166, 0.9, 1.0, 1.3, 1.4, 2.0, 137.9, 31.3, 0.00905, 0, 8, 413, 413, 32, 12700
4 threadCount, timeline, 35761, 1143, 1085, 6727, 1.0, 1.0, 1.4, 1.6, 2.1, 138.0, 31.3, 0.00905, 0, 8, 413, 413, 32, 12700
4 threadCount, total, 142035, 4541, 4368, 10420, 0.8, 0.8, 1.3, 1.4, 1.6, 138.0, 31.3, 0.00905, 0, 8, 413, 413, 32, 12700
8 threadCount, insert, 67924, 2135, 2135, 2924, 0.5, 0.4, 0.7, 0.9, 3.0, 41.6, 31.8, 0.00592, 0, 13, 477, 477, 2, 20864
8 threadCount, singlepost, 137283, 4314, 4098, 4098, 0.9, 1.0, 1.3, 1.5, 1.9, 43.1, 31.8, 0.00592, 0, 13, 477, 477, 2, 20864
8 threadCount, timeline, 68781, 2162, 2052, 12863, 1.1, 1.1, 1.5, 1.7, 2.2, 42.2, 31.8, 0.00592, 0, 13, 477, 477, 2, 20864
8 threadCount, total, 273988, 8611, 8285, 19885, 0.9, 0.8, 1.4, 1.6, 2.0, 43.1, 31.8, 0.00592, 0, 13, 477, 477, 2, 20864
16 threadCount, insert, 120202, 3684, 3684, 5031, 0.6, 0.5, 0.9, 1.1, 1.9, 47.8, 32.6, 0.00506, 0, 22, 872, 872, 3, 35249
16 threadCount, singlepost, 238117, 7297, 6947, 6947, 1.1, 1.1, 1.6, 1.9, 4.1, 49.4, 32.6, 0.00506, 0, 22, 872, 872, 3, 35249
16 threadCount, timeline, 121167, 3713, 3532, 22654, 1.2, 1.2, 1.8, 2.1, 7.1, 49.0, 32.6, 0.00506, 0, 22, 872, 872, 3, 35249
16 threadCount, total, 479486, 14694, 14163, 34631, 1.0, 0.9, 1.6, 1.9, 3.7, 49.4, 32.6, 0.00506, 0, 22, 872, 872, 3, 35249
24 threadCount, insert, 155861, 4687, 4687, 6411, 0.7, 0.6, 1.1, 1.8, 7.0, 70.5, 33.3, 0.00541, 0, 29, 1218, 1262, 6, 47967
24 threadCount, singlepost, 311426, 9364, 8937, 8937, 1.3, 1.2, 2.0, 2.4, 7.2, 71.7, 33.3, 0.00541, 0, 29, 1218, 1262, 6, 47967
24 threadCount, timeline, 156992, 4720, 4502, 29719, 1.5, 1.4, 2.2, 2.6, 42.2, 71.7, 33.3, 0.00541, 0, 29, 1218, 1262, 6, 47967
24 threadCount, total, 624279, 18771, 18125, 45066, 1.2, 1.0, 2.0, 2.3, 6.8, 71.7, 33.3, 0.00541, 0, 29, 1218, 1262, 6, 47967
36 threadCount, insert, 190266, 5468, 5468, 7489, 1.0, 0.8, 1.9, 2.9, 10.6, 267.7, 34.8, 0.00593, 0, 35, 1226, 1417, 3, 56050
36 threadCount, singlepost, 380356, 10930, 10461, 10461, 1.7, 1.5, 2.7, 3.7, 40.5, 57.6, 34.8, 0.00593, 0, 35, 1226, 1417, 3, 56050
36 threadCount, timeline, 189548, 5447, 5211, 35642, 1.9, 1.7, 3.0, 4.3, 44.1, 57.1, 34.8, 0.00593, 0, 35, 1226, 1417, 3, 56050
36 threadCount, total, 760170, 21844, 21140, 53592, 1.6, 1.3, 2.7, 3.5, 9.4, 267.7, 34.8, 0.00593, 0, 35, 1226, 1417, 3, 56050
54 threadCount, insert, 207273, 5642, 5642, 7719, 1.7, 1.2, 3.9, 7.3, 37.8, 56.4, 36.7, 0.00588, 0, 38, 1192, 1503, 2, 60870
54 threadCount, singlepost, 416789, 11344, 10897, 10897, 2.4, 2.0, 4.1, 6.7, 42.7, 56.1, 36.7, 0.00588, 0, 38, 1192, 1503, 2, 60870
54 threadCount, timeline, 206615, 5624, 5403, 38239, 2.6, 2.3, 4.5, 8.0, 43.6, 55.0, 36.7, 0.00588, 0, 38, 1192, 1503, 2, 60870
54 threadCount, total, 830677, 22610, 21942, 56855, 2.3, 1.9, 4.1, 7.1, 42.4, 56.4, 36.7, 0.00588, 0, 38, 1192, 1503, 2, 60870
81 threadCount, insert, 224064, 5377, 5377, 7341, 3.2, 2.3, 6.7, 12.7, 50.5, 350.3, 41.7, 0.00879, 0, 44, 1354, 1883, 5, 72225
81 threadCount, singlepost, 447717, 10745, 10347, 10347, 3.7, 3.1, 6.4, 11.4, 50.4, 291.2, 41.7, 0.00879, 0, 44, 1354, 1883, 5, 72225
81 threadCount, timeline, 224468, 5387, 5188, 37984, 4.0, 3.4, 6.6, 11.3, 50.5, 288.1, 41.7, 0.00879, 0, 44, 1354, 1883, 5, 72225
81 threadCount, total, 896249, 21509, 20913, 55672, 3.7, 3.0, 6.4, 10.6, 49.3, 350.3, 41.7, 0.00879, 0, 44, 1354, 1883, 5, 72225
121 threadCount, insert, 260078, 5322, 5322, 7286, 4.9, 3.8, 11.1, 21.4, 54.6, 166.5, 48.9, 0.00509, 0, 51, 1255, 2034, 2, 81759
121 threadCount, singlepost, 525682, 10758, 10398, 10398, 5.6, 4.9, 9.5, 15.4, 50.4, 151.7, 48.9, 0.00509, 0, 51, 1255, 2034, 2, 81759
121 threadCount, timeline, 260886, 5339, 5157, 39078, 5.9, 5.2, 9.9, 17.1, 51.6, 151.7, 48.9, 0.00509, 0, 51, 1255, 2034, 2, 81759
121 threadCount, total, 1046646, 21419, 20877, 56762, 5.5, 4.8, 9.7, 16.0, 50.6, 166.5, 48.9, 0.00509, 0, 51, 1255, 2034, 2, 81759

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 33438, 1048, 1048, 1436, 0.5, 0.4, 0.7, 0.9, 11.6, 276.4, 31.9, 0.00995, 0, 1, 274, 274, 0, 8124
4 threadCount, singlepost, 66170, 2074, 2007, 2007, 1.0, 1.0, 1.3, 1.4, 1.8, 277.3, 31.9, 0.00995, 0, 1, 274, 274, 0, 8124
4 threadCount, timeline, 33241, 1042, 1008, 7133, 1.1, 1.1, 1.5, 1.6, 2.1, 11.0, 31.9, 0.00995, 0, 1, 274, 274, 0, 8124
4 threadCount, total, 132849, 4164, 4063, 10576, 0.9, 0.9, 1.3, 1.5, 1.9, 277.3, 31.9, 0.00995, 0, 1, 274, 274, 0, 8124
8 threadCount, insert, 61953, 1927, 1927, 2635, 0.5, 0.4, 0.8, 0.9, 11.2, 235.4, 32.1, 0.02825, 0, 3, 700, 700, 0, 24313
8 threadCount, singlepost, 124181, 3863, 3742, 3742, 1.1, 1.1, 1.4, 1.5, 2.1, 236.5, 32.1, 0.02825, 0, 3, 700, 700, 0, 24313
8 threadCount, timeline, 61760, 1921, 1859, 13300, 1.2, 1.2, 1.6, 1.8, 2.3, 236.6, 32.1, 0.02825, 0, 3, 700, 700, 0, 24313
8 threadCount, total, 247894, 7712, 7529, 19677, 1.0, 1.0, 1.4, 1.6, 2.3, 236.6, 32.1, 0.02825, 0, 3, 700, 700, 0, 24313
16 threadCount, insert, 106968, 3253, 3253, 4444, 0.6, 0.5, 0.9, 1.3, 2.2, 243.6, 32.9, 0.01511, 0, 4, 936, 936, 8, 30098
16 threadCount, singlepost, 215904, 6566, 6368, 6368, 1.3, 1.2, 1.7, 2.0, 2.9, 252.0, 32.9, 0.01511, 0, 4, 936, 936, 8, 30098
16 threadCount, timeline, 106719, 3246, 3149, 22826, 1.4, 1.4, 1.9, 2.2, 3.4, 245.1, 32.9, 0.01511, 0, 4, 936, 936, 8, 30098
16 threadCount, total, 429591, 13065, 12770, 33638, 1.1, 1.1, 1.7, 2.0, 2.6, 252.0, 32.9, 0.01511, 0, 4, 936, 936, 8, 30098
24 threadCount, insert, 140739, 4170, 4170, 5704, 0.7, 0.6, 1.3, 1.9, 12.9, 220.6, 33.7, 0.01478, 0, 7, 1334, 1334, 13, 47280
24 threadCount, singlepost, 281951, 8355, 8115, 8115, 1.5, 1.4, 2.0, 2.5, 4.8, 220.0, 33.7, 0.01478, 0, 7, 1334, 1334, 13, 47280
24 threadCount, timeline, 140441, 4162, 4038, 29866, 1.7, 1.6, 2.3, 2.8, 6.7, 219.8, 33.7, 0.01478, 0, 7, 1334, 1334, 13, 47280
24 threadCount, total, 563131, 16686, 16323, 43684, 1.3, 1.3, 2.0, 2.4, 4.5, 220.6, 33.7, 0.01478, 0, 7, 1334, 1334, 13, 47280
36 threadCount, insert, 159910, 4489, 4489, 6153, 1.2, 0.8, 2.4, 4.0, 9.1, 257.0, 35.6, 0.01898, 0, 9, 1717, 1717, 8, 50212
36 threadCount, singlepost, 320744, 9005, 8751, 8751, 2.0, 1.8, 3.1, 4.6, 11.0, 226.2, 35.6, 0.01898, 0, 9, 1717, 1717, 8, 50212
36 threadCount, timeline, 160623, 4509, 4382, 33117, 2.3, 2.0, 3.4, 5.3, 10.1, 211.6, 35.6, 0.01898, 0, 9, 1717, 1717, 8, 50212
36 threadCount, total, 641277, 18003, 17622, 48021, 1.9, 1.7, 3.1, 4.7, 10.8, 257.0, 35.6, 0.01898, 0, 9, 1717, 1717, 8, 50212
54 threadCount, insert, 184306, 4527, 4527, 6189, 2.2, 1.6, 4.6, 8.4, 22.7, 234.8, 40.7, 0.01920, 0, 13, 2092, 2212, 39, 58141
54 threadCount, singlepost, 365911, 8988, 8744, 8744, 3.1, 2.5, 5.0, 8.2, 17.6, 237.7, 40.7, 0.01920, 0, 13, 2092, 2212, 39, 58141
54 threadCount, timeline, 182207, 4476, 4356, 33754, 3.3, 2.8, 5.3, 8.4, 18.1, 237.4, 40.7, 0.01920, 0, 13, 2092, 2212, 39, 58141
54 threadCount, total, 732424, 17991, 17627, 48687, 2.9, 2.4, 5.0, 8.3, 20.4, 237.7, 40.7, 0.01920, 0, 13, 2092, 2212, 39, 58141
81 threadCount, insert, 198077, 4603, 4603, 6296, 3.6, 2.8, 7.0, 11.9, 25.0, 282.6, 43.0, 0.01447, 0, 7, 1702, 1702, 17, 61791
81 threadCount, singlepost, 392856, 9129, 8895, 8895, 4.5, 3.9, 7.5, 11.9, 25.4, 282.4, 43.0, 0.01447, 0, 7, 1702, 1702, 17, 61791
81 threadCount, timeline, 197187, 4582, 4466, 35435, 4.8, 4.2, 7.7, 12.2, 27.2, 282.0, 43.0, 0.01447, 0, 7, 1702, 1702, 17, 61791
81 threadCount, total, 788120, 18315, 17964, 50626, 4.3, 3.8, 7.5, 12.0, 26.4, 282.6, 43.0, 0.01447, 0, 7, 1702, 1702, 17, 61791
121 threadCount, insert, 233633, 4512, 4512, 6159, 5.5, 4.3, 11.0, 18.8, 42.2, 352.3, 51.8, 0.01403, 0, 10, 2035, 2035, 13, 83812
121 threadCount, singlepost, 473282, 9140, 8922, 8922, 6.8, 6.0, 10.7, 16.3, 171.5, 253.0, 51.8, 0.01403, 0, 10, 2035, 2035, 13, 83812
121 threadCount, timeline, 236412, 4566, 4455, 36174, 7.1, 6.4, 11.5, 17.6, 197.3, 256.9, 51.8, 0.01403, 0, 10, 2035, 2035, 13, 83812
121 threadCount, total, 943327, 18218, 17889, 51255, 6.5, 5.8, 10.9, 17.2, 186.2, 352.3, 51.8, 0.01403, 0, 10, 2035, 2035, 13, 83812
181 threadCount, insert, 283444, 4539, 4539, 6209, 6.8, 4.2, 14.1, 20.9, 140.7, 237.1, 62.4, 0.01059, 0, 14, 2545, 2545, 13, 92851
181 threadCount, singlepost, 567229, 9083, 8880, 8880, 10.7, 10.0, 16.5, 24.0, 194.7, 235.9, 62.4, 0.01059, 0, 14, 2545, 2545, 13, 92851
181 threadCount, timeline, 283462, 4539, 4437, 36966, 11.1, 10.3, 16.7, 23.2, 181.4, 235.7, 62.4, 0.01059, 0, 14, 2545, 2545, 13, 92851
181 threadCount, total, 1134135, 18162, 17856, 52054, 9.8, 9.7, 16.0, 23.3, 187.3, 237.1, 62.4, 0.01059, 0, 14, 2545, 2545, 13, 92851

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 29865, 941, 941, 1288, 0.4, 0.4, 0.7, 0.9, 1.3, 24.0, 31.7, 0.01344, 0, 1, 189, 189, 0, 13616
4 threadCount, singlepost, 58981, 1858, 1817, 1817, 1.1, 1.1, 1.4, 1.6, 1.9, 192.1, 31.7, 0.01344, 0, 1, 189, 189, 0, 13616
4 threadCount, timeline, 29198, 920, 902, 7072, 1.3, 1.3, 1.6, 1.8, 2.2, 192.1, 31.7, 0.01344, 0, 1, 189, 189, 0, 13616
4 threadCount, total, 118044, 3718, 3660, 10177, 1.0, 1.1, 1.5, 1.7, 1.9, 192.1, 31.7, 0.01344, 0, 1, 189, 189, 0, 13616
8 threadCount, insert, 54546, 1708, 1708, 2333, 0.5, 0.4, 0.8, 1.0, 2.3, 239.1, 31.9, 0.01014, 0, 2, 396, 396, 33, 27244
8 threadCount, singlepost, 108726, 3404, 3333, 3333, 1.2, 1.2, 1.5, 1.7, 2.2, 243.7, 31.9, 0.01014, 0, 2, 396, 396, 33, 27244
8 threadCount, timeline, 55449, 1736, 1698, 13427, 1.4, 1.4, 1.8, 2.0, 2.4, 240.1, 31.9, 0.01014, 0, 2, 396, 396, 33, 27244
8 threadCount, total, 218721, 6847, 6738, 19092, 1.1, 1.2, 1.6, 1.8, 2.4, 243.7, 31.9, 0.01014, 0, 2, 396, 396, 33, 27244
16 threadCount, insert, 95384, 2838, 2838, 3885, 0.6, 0.5, 1.0, 1.6, 8.5, 294.3, 33.6, 0.01193, 0, 3, 721, 721, 37, 40601
16 threadCount, singlepost, 192030, 5714, 5599, 5599, 1.5, 1.4, 1.9, 2.3, 3.1, 296.1, 33.6, 0.01193, 0, 3, 721, 721, 37, 40601
16 threadCount, timeline, 96518, 2872, 2813, 22426, 1.7, 1.6, 2.1, 2.5, 3.6, 295.6, 33.6, 0.01193, 0, 3, 721, 721, 37, 40601
16 threadCount, total, 383932, 11424, 11249, 31909, 1.3, 1.4, 1.9, 2.2, 3.0, 296.1, 33.6, 0.01193, 0, 3, 721, 721, 37, 40601
24 threadCount, insert, 123912, 3595, 3595, 4927, 0.8, 0.6, 1.5, 2.1, 4.4, 286.7, 34.5, 0.01224, 0, 4, 990, 990, 20, 53495
24 threadCount, singlepost, 244685, 7099, 6955, 6955, 1.8, 1.7, 2.4, 3.0, 4.8, 285.4, 34.5, 0.01224, 0, 4, 990, 990, 20, 53495
24 threadCount, timeline, 122475, 3553, 3485, 28206, 2.0, 1.9, 2.7, 3.3, 7.0, 284.8, 34.5, 0.01224, 0, 4, 990, 990, 20, 53495
24 threadCount, total, 491072, 14248, 14035, 40087, 1.6, 1.6, 2.4, 3.0, 5.4, 286.7, 34.5, 0.01224, 0, 4, 990, 990, 20, 53495
36 threadCount, insert, 141317, 3850, 3850, 5277, 1.3, 1.0, 2.8, 4.8, 16.1, 240.0, 36.7, 0.01231, 0, 4, 913, 913, 7, 52596
36 threadCount, singlepost, 284198, 7743, 7596, 7596, 2.4, 2.2, 3.7, 5.4, 10.0, 243.2, 36.7, 0.01231, 0, 4, 913, 913, 7, 52596
36 threadCount, timeline, 142241, 3876, 3800, 31128, 2.7, 2.4, 3.9, 5.6, 10.1, 244.6, 36.7, 0.01231, 0, 4, 913, 913, 7, 52596
36 threadCount, total, 567756, 15470, 15247, 44000, 2.2, 2.1, 3.7, 5.3, 10.3, 244.6, 36.7, 0.01231, 0, 4, 913, 913, 7, 52596
54 threadCount, insert, 147609, 3469, 3469, 4745, 2.8, 2.2, 5.6, 9.2, 18.7, 287.0, 42.6, 0.02038, 0, 5, 1273, 1273, 19, 61300
54 threadCount, singlepost, 296313, 6963, 6832, 6832, 4.0, 3.5, 6.4, 10.1, 17.9, 289.4, 42.6, 0.02038, 0, 5, 1273, 1273, 19, 61300
54 threadCount, timeline, 148281, 3484, 3420, 28464, 4.3, 3.8, 6.9, 11.0, 19.9, 287.3, 42.6, 0.02038, 0, 5, 1273, 1273, 19, 61300
54 threadCount, total, 592203, 13916, 13721, 40040, 3.8, 3.4, 6.4, 10.1, 18.3, 289.4, 42.6, 0.02038, 0, 5, 1273, 1273, 19, 61300
81 threadCount, insert, 174282, 3778, 3778, 5162, 4.3, 3.5, 8.1, 13.6, 35.7, 338.4, 46.1, 0.01278, 0, 7, 1466, 1466, 10, 77864
81 threadCount, singlepost, 350655, 7600, 7462, 7462, 5.4, 4.9, 8.7, 13.2, 40.4, 248.2, 46.1, 0.01278, 0, 7, 1466, 1466, 10, 77864
81 threadCount, timeline, 176191, 3819, 3750, 31705, 5.8, 5.2, 9.1, 14.1, 25.5, 241.2, 46.1, 0.01278, 0, 7, 1466, 1466, 10, 77864
81 threadCount, total, 701128, 15197, 14990, 44329, 5.2, 4.7, 8.6, 12.2, 24.5, 338.4, 46.1, 0.01278, 0, 7, 1466, 1466, 10, 77864
121 threadCount, insert, 214800, 3841, 3841, 5256, 6.0, 5.1, 12.3, 19.4, 41.9, 230.0, 55.9, 0.01080, 0, 9, 1730, 1730, 8, 87696
121 threadCount, singlepost, 423920, 7580, 7446, 7446, 8.3, 7.8, 13.5, 20.2, 145.0, 248.7, 55.9, 0.01080, 0, 9, 1730, 1730, 8, 87696
121 threadCount, timeline, 211840, 3788, 3721, 31998, 8.7, 8.1, 13.8, 19.8, 47.7, 250.8, 55.9, 0.01080, 0, 9, 1730, 1730, 8, 87696
121 threadCount, total, 850560, 15208, 15008, 44700, 7.8, 7.4, 13.4, 20.0, 47.6, 250.8, 55.9, 0.01080, 0, 9, 1730, 1730, 8, 87696

G1_32G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 30936, 973, 973, 1340, 0.4, 0.4, 0.7, 0.9, 1.4, 17.8, 31.8, 0.00898, 0, 1, 214, 214, 0, 13344
4 threadCount, singlepost, 63324, 1992, 1992, 1992, 1.1, 1.1, 1.3, 1.5, 1.9, 215.8, 31.8, 0.00898, 0, 1, 214, 214, 0, 13344
4 threadCount, timeline, 31379, 987, 987, 6929, 1.2, 1.2, 1.5, 1.7, 2.2, 215.9, 31.8, 0.00898, 0, 1, 214, 214, 0, 13344
4 threadCount, total, 125639, 3952, 3952, 10261, 0.9, 1.0, 1.4, 1.6, 1.9, 215.9, 31.8, 0.00898, 0, 1, 214, 214, 0, 13344
8 threadCount, insert, 59491, 1843, 1843, 2526, 0.5, 0.4, 0.8, 0.9, 1.8, 260.4, 32.3, 0.01028, 0, 2, 446, 446, 36, 27262
8 threadCount, singlepost, 119048, 3688, 3688, 3688, 1.1, 1.1, 1.4, 1.6, 2.0, 261.8, 32.3, 0.01028, 0, 2, 446, 446, 36, 27262
8 threadCount, timeline, 59223, 1835, 1834, 12942, 1.3, 1.3, 1.6, 1.8, 2.1, 261.4, 32.3, 0.01028, 0, 2, 446, 446, 36, 27262
8 threadCount, total, 237762, 7365, 7365, 19155, 1.0, 1.1, 1.5, 1.7, 1.9, 261.8, 32.3, 0.01028, 0, 2, 446, 446, 36, 27262
16 threadCount, insert, 130937, 2899, 2899, 3964, 0.7, 0.5, 1.0, 1.6, 11.6, 282.7, 45.2, 0.01953, 0, 4, 935, 935, 10, 53938
16 threadCount, singlepost, 260371, 5765, 5765, 5765, 1.4, 1.4, 2.0, 2.5, 18.3, 248.8, 45.2, 0.01953, 0, 4, 935, 935, 10, 53938
16 threadCount, timeline, 131599, 2914, 2913, 20924, 1.6, 1.5, 2.2, 2.8, 16.2, 248.8, 45.2, 0.01953, 0, 4, 935, 935, 10, 53938
16 threadCount, total, 522907, 11578, 11577, 30652, 1.3, 1.3, 2.0, 2.5, 15.4, 282.7, 45.2, 0.01953, 0, 4, 935, 935, 10, 53938
24 threadCount, insert, 130883, 3804, 3804, 5202, 0.8, 0.6, 1.4, 2.3, 9.2, 281.2, 34.4, 0.01905, 0, 4, 960, 960, 24, 50744
24 threadCount, singlepost, 260226, 7563, 7563, 7563, 1.6, 1.6, 2.4, 3.2, 7.9, 281.7, 34.4, 0.01905, 0, 4, 960, 960, 24, 50744
24 threadCount, timeline, 131484, 3821, 3821, 28081, 1.9, 1.7, 2.7, 3.5, 8.1, 281.9, 34.4, 0.01905, 0, 4, 960, 960, 24, 50744
24 threadCount, total, 522593, 15189, 15188, 40845, 1.5, 1.4, 2.4, 3.0, 7.9, 281.9, 34.4, 0.01905, 0, 4, 960, 960, 24, 50744
36 threadCount, insert, 188834, 4247, 4247, 5813, 1.3, 0.9, 2.5, 4.1, 10.8, 251.4, 44.5, 0.01940, 0, 7, 1505, 1505, 17, 80632
36 threadCount, singlepost, 376070, 8459, 8458, 8458, 2.2, 1.9, 3.4, 5.1, 14.3, 253.9, 44.5, 0.01940, 0, 7, 1505, 1505, 17, 80632
36 threadCount, timeline, 187315, 4213, 4213, 31700, 2.5, 2.2, 3.7, 5.4, 11.0, 250.7, 44.5, 0.01940, 0, 7, 1505, 1505, 17, 80632
36 threadCount, total, 752219, 16920, 16918, 45970, 2.0, 1.8, 3.4, 5.1, 11.7, 253.9, 44.5, 0.01940, 0, 7, 1505, 1505, 17, 80632
54 threadCount, insert, 171633, 4481, 4481, 6135, 2.2, 1.6, 4.7, 8.2, 13.7, 291.3, 38.3, 0.01210, 0, 6, 1124, 1124, 7, 60444
54 threadCount, singlepost, 345440, 9018, 9017, 9017, 3.1, 2.6, 5.0, 8.2, 13.6, 211.1, 38.3, 0.01210, 0, 6, 1124, 1124, 7, 60444
54 threadCount, timeline, 170465, 4450, 4450, 34329, 3.3, 2.9, 5.5, 9.1, 20.3, 209.1, 38.3, 0.01210, 0, 6, 1124, 1124, 7, 60444
54 threadCount, total, 687538, 17950, 17948, 49481, 2.9, 2.5, 5.1, 8.1, 12.3, 291.3, 38.3, 0.01210, 0, 6, 1124, 1124, 7, 60444
81 threadCount, insert, 332670, 4041, 4041, 5526, 4.1, 3.2, 8.0, 14.0, 59.9, 283.7, 82.3, 0.01953, 0, 14, 2598, 2713, 44, 141155
81 threadCount, singlepost, 669779, 8136, 8135, 8135, 5.0, 4.3, 8.4, 13.5, 41.3, 290.7, 82.3, 0.01953, 0, 14, 2598, 2713, 44, 141155
81 threadCount, timeline, 336680, 4090, 4089, 32559, 5.4, 4.6, 8.8, 14.7, 73.9, 290.4, 82.3, 0.01953, 0, 14, 2598, 2713, 44, 141155
81 threadCount, total, 1339129, 16267, 16266, 46221, 4.9, 4.2, 8.4, 13.6, 47.7, 290.7, 82.3, 0.01953, 0, 14, 2598, 2713, 44, 141155
121 threadCount, insert, 213975, 3926, 3926, 5368, 6.3, 5.1, 12.4, 19.8, 182.6, 419.9, 54.5, 0.01905, 0, 5, 1791, 1791, 38, 89659
121 threadCount, singlepost, 427658, 7846, 7845, 7845, 7.9, 7.2, 12.9, 18.1, 42.6, 413.9, 54.5, 0.01905, 0, 5, 1791, 1791, 38, 89659
121 threadCount, timeline, 212777, 3904, 3903, 32054, 8.3, 7.5, 13.7, 19.2, 43.7, 413.3, 54.5, 0.01905, 0, 5, 1791, 1791, 38, 89659
121 threadCount, total, 854410, 15675, 15674, 45268, 7.6, 6.9, 13.0, 18.3, 37.3, 419.9, 54.5, 0.01905, 0, 5, 1791, 1791, 38, 89659

只有读取

./cassandra-stress user profile=./blogpost1.yaml ops(singlepost=1) -node 10.21.21.10 -log file=SINGLENODE9_blog_q1.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, singlepost, 122443, 4069, 3943, 3943, 0.9, 0.8, 1.5, 1.7, 2.1, 41.4, 30.1, 0.01114, 0, 7, 246, 246, 1, 11380
8 threadCount, singlepost, 235489, 7815, 7565, 7565, 0.9, 0.8, 1.6, 1.8, 2.2, 60.2, 30.1, 0.00628, 0, 12, 426, 426, 2, 19508
16 threadCount, singlepost, 414028, 13749, 13311, 13311, 1.1, 0.9, 1.8, 2.2, 2.8, 49.8, 30.1, 0.00339, 0, 18, 653, 653, 2, 29268
24 threadCount, singlepost, 544879, 18049, 17474, 17474, 1.2, 1.0, 2.2, 2.6, 4.6, 97.9, 30.2, 0.00540, 0, 22, 865, 865, 3, 35767
36 threadCount, singlepost, 656440, 21729, 21036, 21036, 1.6, 1.3, 2.7, 4.0, 10.2, 74.8, 30.2, 0.00489, 0, 26, 959, 959, 1, 42267
54 threadCount, singlepost, 719453, 23782, 23027, 23027, 2.2, 1.7, 3.9, 6.9, 21.9, 87.5, 30.3, 0.01074, 0, 26, 949, 949, 10, 44004
81 threadCount, singlepost, 822082, 26965, 26101, 26101, 2.9, 2.3, 5.5, 11.4, 39.6, 59.5, 30.5, 0.00400, 0, 26, 899, 899, 2, 42271
121 threadCount, singlepost, 881038, 28869, 27948, 27948, 4.1, 3.3, 7.1, 12.4, 41.9, 75.2, 30.5, 0.00428, 0, 26, 912, 912, 1, 42277
181 threadCount, singlepost, 955638, 31123, 30123, 30123, 5.7, 4.8, 9.5, 15.6, 49.8, 108.2, 30.7, 0.00680, 0, 26, 924, 924, 1, 42275
271 threadCount, singlepost, 1010814, 32952, 31906, 31906, 8.1, 7.1, 12.4, 21.1, 51.6, 168.2, 30.7, 0.01138, 0, 25, 907, 907, 2, 40659
406 threadCount, singlepost, 1034735, 33619, 32562, 32562, 12.0, 10.1, 16.9, 36.3, 58.2, 145.2, 30.8, 0.02330, 0, 22, 822, 822, 1, 35778
609 threadCount, singlepost, 1059253, 34225, 33141, 33141, 17.7, 16.2, 23.2, 57.2, 79.0, 219.8, 30.9, 0.01185, 0, 22, 825, 825, 1, 35796
913 threadCount, singlepost, 1029097, 32958, 31916, 31916, 27.6, 25.9, 31.3, 66.4, 163.6, 309.2, 31.2, 0.01027, 0, 21, 777, 777, 2, 34169

G1_16G

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, singlepost, 109941, 3639, 3560, 3560, 1.0, 1.0, 1.4, 1.6, 1.8, 178.4, 30.2, 0.00904, 0, 2, 320, 320, 16, 11256
8 threadCount, singlepost, 207401, 6893, 6743, 6743, 1.1, 1.1, 1.4, 1.6, 1.9, 145.8, 30.1, 0.00604, 0, 3, 422, 422, 3, 16829
16 threadCount, singlepost, 354764, 11775, 11517, 11517, 1.3, 1.2, 1.7, 2.0, 2.6, 139.1, 30.1, 0.00768, 0, 6, 800, 800, 3, 33620
24 threadCount, singlepost, 448907, 14829, 14506, 14506, 1.5, 1.4, 2.2, 2.7, 4.6, 143.3, 30.3, 0.01183, 0, 7, 880, 880, 16, 39657
36 threadCount, singlepost, 492859, 16296, 15946, 15946, 2.1, 1.9, 3.4, 5.6, 10.4, 104.4, 30.2, 0.00672, 0, 7, 589, 589, 6, 40129
54 threadCount, singlepost, 503342, 16622, 16262, 16262, 3.2, 2.8, 5.2, 8.5, 15.2, 103.2, 30.3, 0.00200, 0, 8, 597, 597, 3, 45974
81 threadCount, singlepost, 516925, 17020, 16647, 16647, 4.7, 4.2, 7.7, 12.4, 78.0, 99.8, 30.4, 0.00587, 0, 7, 518, 518, 2, 40218
121 threadCount, singlepost, 525550, 17186, 16816, 16816, 6.9, 6.2, 11.5, 16.5, 82.1, 111.6, 30.6, 0.00364, 0, 8, 588, 588, 2, 45943
181 threadCount, singlepost, 544759, 17741, 17353, 17353, 10.1, 9.2, 15.7, 20.7, 87.8, 120.6, 30.7, 0.00363, 0, 8, 596, 596, 2, 45887
271 threadCount, singlepost, 551750, 17863, 17477, 17477, 15.1, 14.0, 22.4, 27.7, 96.8, 144.2, 30.9, 0.00352, 0, 8, 587, 587, 3, 45895

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, singlepost, 92111, 3061, 3008, 3008, 1.2, 1.2, 1.5, 1.6, 1.9, 166.0, 30.1, 0.00286, 0, 1, 163, 163, 0, 8953
8 threadCount, singlepost, 173858, 5753, 5654, 5654, 1.3, 1.3, 1.6, 1.8, 2.0, 137.5, 30.2, 0.00164, 0, 2, 268, 268, 0, 18140
16 threadCount, singlepost, 285754, 9470, 9312, 9312, 1.6, 1.5, 2.1, 2.5, 3.3, 141.4, 30.2, 0.00497, 0, 4, 545, 545, 2, 36332
24 threadCount, singlepost, 339980, 11279, 11079, 11079, 2.0, 1.9, 2.9, 3.6, 6.2, 154.3, 30.1, 0.00245, 0, 5, 701, 701, 5, 45509
36 threadCount, singlepost, 364567, 12065, 11862, 11862, 2.9, 2.6, 4.4, 6.7, 11.8, 125.5, 30.2, 0.00387, 0, 5, 469, 469, 16, 46767
54 threadCount, singlepost, 374115, 12349, 12135, 12135, 4.3, 4.0, 6.9, 10.4, 16.1, 96.2, 30.3, 0.00199, 0, 5, 382, 382, 2, 46894
81 threadCount, singlepost, 390179, 12834, 12610, 12610, 6.2, 5.9, 10.3, 15.3, 26.7, 112.0, 30.4, 0.00922, 0, 5, 395, 395, 3, 46911
121 threadCount, singlepost, 412920, 13492, 13259, 13259, 8.9, 8.1, 14.5, 19.9, 33.7, 123.5, 30.6, 0.00379, 0, 5, 391, 391, 2, 46859
181 threadCount, singlepost, 439554, 14237, 13989, 13989, 12.6, 11.7, 19.8, 25.0, 94.3, 139.9, 30.9, 0.00353, 0, 6, 465, 465, 2, 56229
271 threadCount, singlepost, 459036, 14783, 14527, 14527, 18.2, 17.3, 26.3, 31.4, 107.9, 150.9, 31.1, 0.00419, 0, 5, 386, 386, 2, 46835
406 threadCount, singlepost, 472232, 15120, 14857, 14857, 26.7, 25.4, 34.9, 47.8, 110.0, 147.2, 31.2, 0.00490, 0, 6, 473, 473, 2, 56115
609 threadCount, singlepost, 495949, 15634, 15357, 15357, 38.9, 37.7, 48.3, 57.1, 213.9, 238.0, 31.7, 0.00602, 0, 5, 400, 400, 3, 46782
913 threadCount, singlepost, 509520, 15849, 15572, 15572, 57.6, 53.7, 72.2, 138.2, 210.9, 245.9, 32.1, 0.00958, 0, 6, 480, 480, 3, 56182

G1_32G

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, singlepost, 99639, 3285, 3285, 3285, 1.1, 1.1, 1.6, 2.0, 2.6, 285.7, 30.3, 0.00463, 0, 1, 283, 283, 0, 17966
8 threadCount, singlepost, 190521, 6331, 6331, 6331, 1.2, 1.1, 1.7, 2.0, 2.6, 238.7, 30.1, 0.00805, 0, 1, 235, 235, 0, 18336
16 threadCount, singlepost, 330162, 10888, 10888, 10888, 1.4, 1.3, 2.0, 2.3, 3.0, 234.4, 30.3, 0.00838, 0, 2, 436, 436, 13, 36640
24 threadCount, singlepost, 422433, 13919, 13918, 13918, 1.6, 1.5, 2.5, 3.1, 4.7, 229.7, 30.3, 0.00868, 0, 2, 444, 444, 3, 36672
36 threadCount, singlepost, 495485, 16338, 16337, 16337, 2.1, 1.9, 3.4, 5.1, 8.5, 219.4, 30.3, 0.00704, 0, 3, 629, 629, 4, 54968
54 threadCount, singlepost, 540856, 17748, 17746, 17746, 3.0, 2.5, 5.0, 8.2, 13.3, 235.9, 30.5, 0.00533, 0, 3, 655, 655, 7, 54933
81 threadCount, singlepost, 542694, 17868, 17866, 17866, 4.4, 3.9, 7.6, 12.3, 20.1, 225.0, 30.4, 0.00681, 0, 2, 381, 381, 17, 37201
121 threadCount, singlepost, 572196, 18786, 18785, 18785, 6.3, 5.4, 11.5, 17.8, 31.0, 155.7, 30.5, 0.01925, 0, 3, 349, 349, 21, 57439
181 threadCount, singlepost, 642455, 20933, 20931, 20931, 8.5, 7.7, 13.5, 18.2, 36.2, 119.7, 30.7, 0.00638, 0, 3, 228, 228, 2, 58005
271 threadCount, singlepost, 653435, 21239, 21237, 21237, 12.7, 11.7, 18.8, 23.3, 86.6, 162.8, 30.8, 0.01504, 0, 3, 226, 226, 2, 58039
406 threadCount, singlepost, 651626, 20891, 20889, 20889, 19.3, 18.1, 26.6, 31.8, 135.3, 157.1, 31.2, 0.00646, 0, 3, 224, 224, 1, 58031
609 threadCount, singlepost, 649846, 20818, 20816, 20816, 29.2, 28.4, 39.0, 49.2, 140.8, 168.4, 31.2, 0.00658, 0, 3, 220, 220, 1, 58031

./cassandra-stress user profile=./blogpost1.yaml ops(timeline=1) -node 10.21.21.10 -log file=SINGLENODE10_blog_q2.log

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, timeline, 141365, 4698, 4551, 35081, 0.8, 0.8, 1.0, 1.2, 1.3, 29.3, 30.1, 0.00663, 0, 7, 180, 180, 1, 11430
8 threadCount, timeline, 265718, 8824, 8538, 65799, 0.8, 0.8, 1.1, 1.2, 1.6, 46.5, 30.1, 0.00213, 0, 12, 322, 322, 2, 19593
16 threadCount, timeline, 474596, 15761, 15255, 117608, 0.9, 0.9, 1.2, 1.4, 2.2, 30.6, 30.1, 0.00551, 0, 20, 526, 526, 1, 32658
24 threadCount, timeline, 636296, 21118, 20452, 157511, 1.0, 1.0, 1.4, 1.7, 3.5, 38.3, 30.1, 0.00725, 0, 26, 698, 698, 2, 42462
36 threadCount, timeline, 783994, 25984, 25153, 193866, 1.3, 1.2, 2.0, 2.9, 10.2, 32.2, 30.2, 0.00573, 0, 32, 804, 858, 1, 52263
54 threadCount, timeline, 829706, 27400, 26524, 204420, 1.9, 1.5, 3.3, 6.3, 14.7, 43.7, 30.3, 0.00411, 0, 33, 805, 884, 1, 53898
81 threadCount, timeline, 845373, 27903, 27028, 208271, 2.8, 2.2, 5.1, 9.8, 31.3, 59.9, 30.3, 0.00598, 0, 33, 815, 897, 1, 53902
121 threadCount, timeline, 867840, 28402, 27500, 211936, 4.2, 3.3, 7.4, 13.9, 35.8, 74.6, 30.6, 0.00515, 0, 34, 814, 920, 2, 55536
181 threadCount, timeline, 864168, 28392, 27486, 211904, 6.3, 5.6, 10.1, 15.0, 36.1, 109.1, 30.4, 0.00668, 0, 33, 792, 868, 1, 53909
271 threadCount, timeline, 850865, 27816, 26934, 207565, 9.6, 8.9, 14.4, 20.4, 45.8, 168.7, 30.6, 0.00722, 0, 31, 796, 822, 1, 50649

G1_16G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, timeline, 94817, 3150, 3081, 26009, 1.2, 1.2, 1.5, 1.7, 1.9, 79.5, 30.1, 0.01005, 0, 3, 226, 226, 2, 17204
8 threadCount, timeline, 180270, 5989, 5857, 49467, 1.3, 1.3, 1.6, 1.8, 2.1, 77.2, 30.1, 0.00188, 0, 4, 285, 285, 2, 22916
16 threadCount, timeline, 309748, 10283, 10060, 84849, 1.5, 1.4, 1.9, 2.2, 2.8, 78.7, 30.1, 0.00332, 0, 7, 485, 485, 3, 40163
24 threadCount, timeline, 390132, 12948, 12665, 106703, 1.8, 1.7, 2.5, 3.1, 5.9, 74.7, 30.1, 0.00768, 0, 8, 525, 525, 3, 45882
36 threadCount, timeline, 409718, 13539, 13247, 111801, 2.6, 2.3, 4.0, 6.0, 10.0, 80.8, 30.3, 0.00283, 0, 9, 597, 597, 2, 51530
54 threadCount, timeline, 419480, 13781, 13479, 113650, 3.8, 3.5, 6.1, 9.2, 69.8, 81.4, 30.4, 0.00249, 0, 9, 597, 597, 1, 51606
81 threadCount, timeline, 424742, 13963, 13663, 115244, 5.7, 5.4, 9.1, 13.3, 72.4, 98.2, 30.4, 0.00410, 0, 10, 666, 666, 2, 57182
121 threadCount, timeline, 453080, 14847, 14527, 122499, 8.0, 7.4, 12.9, 17.5, 48.8, 98.5, 30.5, 0.00354, 0, 9, 601, 601, 2, 51426
181 threadCount, timeline, 484441, 15578, 15240, 128493, 11.5, 10.4, 18.1, 24.0, 84.9, 121.2, 31.1, 0.00916, 0, 10, 658, 658, 2, 57159
271 threadCount, timeline, 497398, 16078, 15732, 132660, 16.7, 15.4, 25.0, 32.6, 88.4, 157.2, 30.9, 0.00627, 0, 10, 648, 648, 2, 57157
406 threadCount, timeline, 495006, 15889, 15550, 131065, 25.4, 24.3, 36.4, 65.2, 107.0, 139.0, 31.2, 0.00792, 0, 10, 662, 662, 2, 57141
609 threadCount, timeline, 494388, 15750, 15409, 129898, 38.6, 39.5, 51.2, 65.6, 316.1, 328.2, 31.4, 0.01020, 0, 9, 602, 602, 3, 51304
913 threadCount, timeline, 513498, 16183, 15834, 133546, 56.3, 54.7, 74.1, 123.2, 181.0, 234.1, 31.7, 0.01149, 0, 9, 595, 595, 5, 51276

G1_24G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, timeline, 92842, 3085, 3033, 26250, 1.2, 1.3, 1.7, 1.8, 2.1, 80.6, 30.1, 0.00626, 0, 1, 78, 78, 0, 9368
8 threadCount, timeline, 178117, 5917, 5817, 50390, 1.3, 1.4, 1.8, 2.0, 2.3, 80.6, 30.1, 0.00197, 0, 3, 229, 229, 2, 28092
16 threadCount, timeline, 312579, 10362, 10186, 88208, 1.5, 1.5, 2.1, 2.4, 3.2, 82.0, 30.2, 0.00372, 0, 4, 301, 301, 2, 37464
24 threadCount, timeline, 399572, 13245, 13017, 112750, 1.7, 1.8, 2.6, 3.3, 5.4, 83.3, 30.2, 0.00266, 0, 6, 447, 447, 2, 56207
36 threadCount, timeline, 439440, 14540, 14291, 123743, 2.4, 2.2, 4.0, 6.0, 10.2, 90.0, 30.2, 0.00303, 0, 5, 374, 374, 2, 46855
54 threadCount, timeline, 486527, 13926, 13693, 118602, 3.8, 3.4, 6.6, 9.7, 17.6, 916.4, 34.9, 0.01927, 0, 6, 444, 444, 5, 56223
81 threadCount, timeline, 454010, 14970, 14714, 127580, 5.3, 4.8, 9.2, 14.0, 39.8, 175.2, 30.3, 0.01782, 0, 6, 485, 485, 7, 56195
121 threadCount, timeline, 503869, 16544, 16259, 140877, 7.2, 6.6, 12.3, 17.0, 81.2, 106.9, 30.5, 0.00660, 0, 6, 440, 440, 2, 56155
181 threadCount, timeline, 544709, 17747, 17446, 151122, 10.1, 9.3, 15.8, 21.1, 87.0, 131.8, 30.7, 0.00631, 0, 6, 426, 426, 2, 56127
271 threadCount, timeline, 569370, 18463, 18148, 157275, 14.6, 13.5, 22.3, 26.8, 89.7, 149.3, 30.8, 0.00454, 0, 6, 463, 463, 15, 56122
406 threadCount, timeline, 560757, 17992, 17679, 153140, 22.4, 21.9, 32.9, 42.1, 133.9, 198.9, 31.2, 0.01142, 0, 6, 414, 414, 3, 56115
609 threadCount, timeline, 579726, 18491, 18181, 157507, 32.8, 32.1, 44.9, 59.0, 141.0, 177.7, 31.4, 0.00779, 0, 6, 421, 421, 6, 56091
913 threadCount, timeline, 609238, 19220, 18888, 163580, 47.4, 45.8, 58.4, 111.9, 283.9, 306.1, 31.7, 0.01073, 0, 6, 414, 414, 2, 56103

G1_32G

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, timeline, 106287, 3532, 3532, 29277, 1.1, 1.1, 1.4, 1.5, 1.7, 77.9, 30.1, 0.00395, 0, 1, 76, 76, 0, 19328
8 threadCount, timeline, 200907, 6677, 6676, 55365, 1.1, 1.1, 1.4, 1.6, 1.8, 79.5, 30.1, 0.00313, 0, 1, 76, 76, 0, 19335
16 threadCount, timeline, 350519, 11640, 11639, 96518, 1.3, 1.3, 1.7, 1.9, 2.6, 82.9, 30.1, 0.01155, 0, 3, 224, 224, 3, 58033
24 threadCount, timeline, 454119, 15074, 15073, 124972, 1.5, 1.4, 2.1, 2.6, 4.1, 83.0, 30.1, 0.00830, 0, 3, 225, 225, 4, 58022
36 threadCount, timeline, 485724, 16081, 16080, 133276, 2.1, 1.9, 3.4, 5.2, 8.3, 80.1, 30.2, 0.00793, 0, 3, 216, 216, 0, 58039
54 threadCount, timeline, 493962, 16307, 16306, 135098, 3.2, 2.9, 5.3, 8.8, 14.4, 90.9, 30.3, 0.00459, 0, 3, 223, 223, 1, 58007
81 threadCount, timeline, 496295, 16383, 16382, 135850, 4.8, 4.4, 7.9, 11.7, 22.5, 114.7, 30.3, 0.00669, 0, 4, 293, 293, 1, 77393
121 threadCount, timeline, 509577, 16710, 16708, 138473, 7.1, 6.4, 11.6, 15.8, 34.3, 92.5, 30.5, 0.00294, 0, 3, 215, 215, 1, 58039
181 threadCount, timeline, 526174, 17130, 17129, 142058, 10.5, 9.6, 16.1, 20.2, 33.4, 114.9, 30.7, 0.00473, 0, 4, 295, 295, 2, 77407
271 threadCount, timeline, 530603, 17213, 17212, 142632, 15.6, 14.6, 23.7, 29.1, 97.5, 154.6, 30.8, 0.00926, 0, 3, 220, 220, 3, 58047

G1测试

1
2
3
4
5
#/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress write n=50000000 truncate=always -node 10.21.21.10 
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress write n=50000000 cl=ONE truncate=always -mode native cql3 -node 10.21.21.10
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress read n=200000 no-warmup -rate threads\>=16 -node 10.21.21.10
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=1,read=3\) n=100000 cl=ONE -pop dist=UNIFORM\(1..1000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -node 10.21.21.10
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=3,read=1\) n=100000 cl=ONE -pop dist=UNIFORM\(1..1000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -node 10.21.21.10

CMS Heap/New=8G/2G

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Results:
op rate : 126219 [WRITE:126219]
partition rate : 126219 [WRITE:126219]
row rate : 126219 [WRITE:126219]
latency mean : 1.6 [WRITE:1.6]
latency median : 1.3 [WRITE:1.3]
latency 95th percentile : 1.9 [WRITE:1.9]
latency 99th percentile : 3.0 [WRITE:3.0]
latency 99.9th percentile : 19.2 [WRITE:19.2]
latency max : 5649.4 [WRITE:5649.4]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 335
total gc mb : 486236
total gc time (s) : 41
avg gc time(ms) : 121
stdev gc time(ms) : 223
Total operation time : 00:06:36

G1 Heap=16G(16384M),或者微调G1的选项,设置目标GC时间为1s,设置当堆内存占用60%时触发执行GC。

1
2
3
#JVM_OPTS="$JVM_OPTS -XX:MaxGCPauseMillis=500"
JVM_OPTS="$JVM_OPTS -XX:MaxGCPauseMillis=1000"
JVM_OPTS="$JVM_OPTS -XX:InitiatingHeapOccupancyPercent=60"

测试结果:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 117340 [WRITE:117340]
partition rate : 117340 [WRITE:117340]
row rate : 117340 [WRITE:117340]
latency mean : 1.7 [WRITE:1.7]
latency median : 1.3 [WRITE:1.3]
latency 95th percentile : 2.7 [WRITE:2.7]
latency 99th percentile : 4.1 [WRITE:4.1]
latency 99.9th percentile : 13.2 [WRITE:13.2]
latency max : 654.6 [WRITE:654.6]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 102
total gc mb : 508991
total gc time (s) : 23
avg gc time(ms) : 226
stdev gc time(ms) : 80
Total operation time : 00:07:06

G1 Heap=24G(24576M)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 134586 [WRITE:134586]
partition rate : 134586 [WRITE:134586]
row rate : 134586 [WRITE:134586]
latency mean : 1.5 [WRITE:1.5]
latency median : 1.3 [WRITE:1.3]
latency 95th percentile : 1.6 [WRITE:1.6]
latency 99th percentile : 2.5 [WRITE:2.5]
latency 99.9th percentile : 6.4 [WRITE:6.4]
latency max : 637.7 [WRITE:637.7]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 60
total gc mb : 438681
total gc time (s) : 21
avg gc time(ms) : 346
stdev gc time(ms) : 102
Total operation time : 00:06:11

JDK: zulu8.15.0.1-jdk8.0.92-linux_x64
测试结果显示,zulu在24G下并没有性能提升,相反每秒钟的吞吐量下降了2万多,GC也好不到哪里去。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 111652 [WRITE:111652]
partition rate : 111652 [WRITE:111652]
row rate : 111652 [WRITE:111652]
latency mean : 1.8 [WRITE:1.8]
latency median : 1.4 [WRITE:1.4]
latency 95th percentile : 2.9 [WRITE:2.9]
latency 99th percentile : 4.5 [WRITE:4.5]
latency 99.9th percentile : 11.8 [WRITE:11.8]
latency max : 1820.3 [WRITE:1820.3]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 67
total gc mb : 435180
total gc time (s) : 23
avg gc time(ms) : 344
stdev gc time(ms) : 146
Total operation time : 00:07:27

G1 Heap=32G(32768M)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 120814 [WRITE:120814]
partition rate : 120814 [WRITE:120814]
row rate : 120814 [WRITE:120814]
latency mean : 1.6 [WRITE:1.6]
latency median : 1.3 [WRITE:1.3]
latency 95th percentile : 2.4 [WRITE:2.4]
latency 99th percentile : 3.9 [WRITE:3.9]
latency 99.9th percentile : 11.1 [WRITE:11.1]
latency max : 1622.8 [WRITE:1622.8]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 51
total gc mb : 567608
total gc time (s) : 22
avg gc time(ms) : 429
stdev gc time(ms) : 144
Total operation time : 00:06:53

9个节点的集群: /home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress write n=50000000 cl=ONE truncate=always -mode native cql3 -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
op rate                   : 308252 [WRITE:308252]
partition rate : 308252 [WRITE:308252]
row rate : 308252 [WRITE:308252]
latency mean : 0.6 [WRITE:0.6]
latency median : 0.3 [WRITE:0.3]
latency 95th percentile : 2.1 [WRITE:2.1]
latency 99th percentile : 3.2 [WRITE:3.2]
latency 99.9th percentile : 6.2 [WRITE:6.2]
latency max : 495.6 [WRITE:495.6]
Total partitions : 50000000 [WRITE:50000000]
Total errors : 0 [WRITE:0]
total gc count : 83
total gc mb : 633231
total gc time (s) : 14
avg gc time(ms) : 165
stdev gc time(ms) : 101
Total operation time : 00:02:42

集群测试

10.21.21.10,10.21.21.11,10.21.21.12,10.21.21.13,10.21.21.14,10.21.21.15,10.21.21.16,10.21.21.17,10.21.21.18
10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130

nohup sh stress1.sh > stress1.log 2>&1 &

100万写,20万读,10万混合读写;5000万写,100万混合读写;自定义表结构

1
2
3
4
5
6
7
8
9
10
11
12
13
14
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress write n=1000000 -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER401_write_10M.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress read n=200000 -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER402_read_2.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress read n=200000 no-warmup -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER403_read_2_nowarm.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress read duration=3m -rate threads\>=16 -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER404_read_3min.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=1,read=3\) n=100000 cl=ONE -pop dist=UNIFORM\(1..10000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -log file=CLUSTER405_mixed_read3_10M.log -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=3,read=1\) n=100000 cl=ONE -pop dist=UNIFORM\(1..10000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -log file=CLUSTER406_mixed_write3_10M.log -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress write n=50000000 cl=ONE -mode native cql3 -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER407_write_50M.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=1,read=3\) n=1000000 cl=ONE -pop dist=UNIFORM\(1..10000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -log file=CLUSTER408_mixed_read3_100M.log -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress mixed ratio\(write=3,read=1\) n=1000000 cl=ONE -pop dist=UNIFORM\(1..10000000\) -schema keyspace="keyspace1" -mode native cql3 -rate threads\>=16 -log file=CLUSTER409_mixed_write3_100M.log -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130

/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress user profile=blogpost3.yaml ops\(insert=1\) -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER410_blog_insert.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress user profile=blogpost3.yaml ops\(singlepost=2,timeline=1,insert=1\) -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER411_blog_mixed.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress user profile=blogpost3.yaml ops\(singlepost=1\) -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER412_blog_q1.log
/home/admin/apache-cassandra-2.2.6/tools/bin/cassandra-stress user profile=blogpost3.yaml ops\(timeline=1\) -node 10.21.21.19,10.21.21.20,10.21.21.21,10.21.21.22,10.21.21.23,10.21.21.24,10.21.21.25,10.21.21.26,10.21.21.130 -log file=CLUSTER413_blog_q2.log

CLUSTER41_write_1M.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Results:
op rate : 259897 [WRITE:259897]
partition rate : 259897 [WRITE:259897]
row rate : 259897 [WRITE:259897]
latency mean : 0.8 [WRITE:0.8]
latency median : 0.5 [WRITE:0.5]
latency 95th percentile : 1.4 [WRITE:1.4]
latency 99th percentile : 2.0 [WRITE:2.0]
latency 99.9th percentile : 30.7 [WRITE:30.7]
latency max : 95.2 [WRITE:95.2]
Total partitions : 1000000 [WRITE:1000000]
Total errors : 0 [WRITE:0]
total gc count : 4
total gc mb : 6166
total gc time (s) : 0
avg gc time(ms) : 79
stdev gc time(ms) : 9
Total operation time : 00:00:03

CLUSTER42_read_2.log

1
2
3
4
5
6
7
8
9
10
11
12
13
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 21460, 21460, 21460, 0.2, 0.2, 0.2, 0.3, 0.5, 10.6, 9.3, 0.00255, 0, 0, 0, 0, 0, 0
8 threadCount, READ, 200000, 38502, 38502, 38502, 0.2, 0.2, 0.3, 0.3, 0.6, 140.3, 5.2, 0.00191, 0, 1, 139, 139, 0, 1547
16 threadCount, READ, 200000, 71834, 71834, 71834, 0.2, 0.2, 0.3, 0.4, 0.6, 16.0, 2.8, 0.00155, 0, 0, 0, 0, 0, 0
24 threadCount, READ, 200000, 83062, 83062, 83062, 0.3, 0.2, 0.3, 0.6, 1.3, 121.3, 2.4, 0.04126, 0, 3, 349, 349, 3, 4624
36 threadCount, READ, 200000, 134469, 134469, 134469, 0.3, 0.2, 0.4, 0.5, 0.7, 3.7, 1.5, 0.00428, 0, 0, 0, 0, 0, 0
54 threadCount, READ, 200000, 162900, 162900, 162900, 0.3, 0.3, 0.5, 0.6, 0.8, 40.7, 1.2, 0.01089, 0, 1, 39, 39, 0, 1678
81 threadCount, READ, 200000, 216457, 216457, 216457, 0.4, 0.3, 0.6, 0.8, 1.5, 5.8, 0.9, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 200000, 236040, 236040, 236040, 0.5, 0.4, 0.9, 1.2, 14.8, 40.8, 0.8, 0.00000, 0, 2, 26, 26, 0, 3428
181 threadCount, READ, 200000, 286542, 286542, 286542, 0.6, 0.5, 1.2, 1.6, 1.8, 16.6, 0.7, 0.00000, 0, 1, 12, 12, 0, 1713
271 threadCount, READ, 200000, 299832, 299832, 299832, 0.9, 0.7, 2.0, 2.4, 3.5, 45.7, 0.7, 0.00000, 0, 0, 0, 0, 0, 0
406 threadCount, READ, 200000, 321774, 321774, 321774, 1.3, 1.0, 3.2, 4.2, 6.9, 20.1, 0.6, 0.00000, 0, 1, 12, 12, 0, 1662
609 threadCount, READ, 200000, 266682, 266682, 266682, 2.4, 1.2, 6.3, 14.6, 84.6, 88.3, 0.7, 0.00000, 0, 1, 12, 12, 0, 1659

CLUSTER43_read_2_nowarm.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 200000, 20221, 20221, 20221, 0.2, 0.2, 0.2, 0.3, 0.8, 14.0, 9.9, 0.02496, 0, 1, 12, 12, 0, 1658
8 threadCount, READ, 200000, 39354, 39354, 39354, 0.2, 0.2, 0.3, 0.3, 0.5, 15.9, 5.1, 0.00334, 0, 1, 13, 13, 0, 1657
16 threadCount, READ, 200000, 69779, 69779, 69779, 0.2, 0.2, 0.3, 0.4, 0.8, 14.0, 2.9, 0.00565, 0, 1, 13, 13, 0, 1643
24 threadCount, READ, 200000, 95817, 95817, 95817, 0.2, 0.2, 0.3, 0.5, 0.9, 14.0, 2.1, 0.00720, 0, 1, 12, 12, 0, 1643
36 threadCount, READ, 200000, 130448, 130448, 130448, 0.3, 0.2, 0.4, 0.6, 1.9, 19.1, 1.5, 0.00538, 0, 1, 13, 13, 0, 1642
54 threadCount, READ, 200000, 161641, 161641, 161641, 0.3, 0.3, 0.5, 0.6, 1.0, 41.0, 1.2, 0.00939, 0, 1, 13, 13, 0, 1640
81 threadCount, READ, 200000, 207644, 207644, 207644, 0.4, 0.3, 0.6, 0.8, 1.2, 14.4, 1.0, 0.00000, 0, 1, 13, 13, 0, 1637
121 threadCount, READ, 200000, 225275, 225275, 225275, 0.5, 0.4, 0.9, 1.2, 32.9, 65.8, 0.9, 0.00000, 0, 1, 13, 13, 0, 1638
181 threadCount, READ, 200000, 252569, 252569, 252569, 0.7, 0.5, 1.4, 1.7, 2.3, 80.4, 0.8, 0.00000, 0, 0, 0, 0, 0, 0
271 threadCount, READ, 200000, 263521, 263521, 263521, 1.0, 0.7, 1.9, 2.8, 14.5, 71.3, 0.8, 0.00000, 0, 1, 12, 12, 0, 1640
406 threadCount, READ, 200000, 301401, 301401, 301401, 1.4, 1.0, 3.2, 5.1, 52.4, 53.6, 0.7, 0.00000, 0, 1, 12, 12, 0, 1640
609 threadCount, READ, 200000, 298018, 298018, 298018, 2.1, 1.2, 5.9, 9.9, 15.4, 35.7, 0.7, 0.00000, 0, 2, 26, 26, 0, 3280
913 threadCount, READ, 200000, 285031, 285031, 285031, 3.4, 1.2, 11.4, 14.0, 36.6, 50.0, 0.7, 0.00000, 0, 0, 0, 0, 0, 0

CLUSTER44_read_3min.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, READ, 3899838, 21666, 21666, 21666, 0.2, 0.2, 0.2, 0.3, 0.3, 17.5, 180.0, 0.00060, 0, 20, 264, 264, 1, 32759
8 threadCount, READ, 7168561, 39825, 39825, 39825, 0.2, 0.2, 0.2, 0.3, 0.4, 22.0, 180.0, 0.00043, 0, 36, 477, 477, 2, 58983
16 threadCount, READ, 12822843, 71238, 71238, 71238, 0.2, 0.2, 0.3, 0.4, 0.6, 21.8, 180.0, 0.00075, 0, 59, 760, 760, 2, 96665
24 threadCount, READ, 17709667, 98387, 98387, 98387, 0.2, 0.2, 0.3, 0.4, 0.8, 21.2, 180.0, 0.00096, 0, 79, 980, 980, 1, 129439
36 threadCount, READ, 23887360, 132707, 132707, 132707, 0.3, 0.2, 0.4, 0.5, 0.9, 21.3, 180.0, 0.00101, 0, 105, 1314, 1314, 1, 172032
54 threadCount, READ, 31144699, 173025, 173025, 173025, 0.3, 0.3, 0.4, 0.6, 1.2, 27.8, 180.0, 0.00124, 0, 135, 1654, 1654, 1, 221187
81 threadCount, READ, 38692768, 214959, 214959, 214959, 0.4, 0.3, 0.6, 0.7, 1.5, 40.2, 180.0, 0.00126, 0, 166, 2010, 2010, 1, 271986
121 threadCount, READ, 46099747, 256109, 256109, 256109, 0.5, 0.4, 0.8, 1.0, 1.7, 42.0, 180.0, 0.00132, 0, 195, 2352, 2352, 1, 319494
181 threadCount, READ, 51810325, 287834, 287834, 287834, 0.6, 0.5, 1.2, 1.5, 2.4, 51.7, 180.0, 0.00156, 0, 218, 2583, 2583, 1, 357188
271 threadCount, READ, 56578917, 314325, 314325, 314325, 0.8, 0.7, 1.7, 2.3, 3.9, 44.5, 180.0, 0.00163, 0, 238, 2818, 2830, 1, 388317
406 threadCount, READ, 59909240, 332826, 332826, 332826, 1.2, 0.9, 2.7, 4.0, 13.9, 208.3, 180.0, 0.00223, 0, 252, 2976, 3000, 1, 409622
609 threadCount, READ, 56825414, 315692, 315692, 315692, 1.9, 1.0, 5.4, 7.1, 15.6, 47.6, 180.0, 0.00207, 0, 238, 2826, 2826, 1, 389955
913 threadCount, READ, 53999369, 299986, 299986, 299986, 3.0, 1.3, 8.9, 11.0, 15.6, 302.9, 180.0, 0.00160, 0, 225, 2755, 2755, 1, 368647

CLUSTER45_write_1M_rows.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Results:
op rate : 267419 [WRITE:267419]
partition rate : 267419 [WRITE:267419]
row rate : 267419 [WRITE:267419]
latency mean : 0.7 [WRITE:0.7]
latency median : 0.5 [WRITE:0.5]
latency 95th percentile : 1.2 [WRITE:1.2]
latency 99th percentile : 1.8 [WRITE:1.8]
latency 99.9th percentile : 31.9 [WRITE:31.9]
latency max : 218.2 [WRITE:218.2]
Total partitions : 1000000 [WRITE:1000000]
Total errors : 0 [WRITE:0]
total gc count : 4
total gc mb : 6248
total gc time (s) : 0
avg gc time(ms) : 31
stdev gc time(ms) : 11
Total operation time : 00:00:03

CLUSTER46_mixed_autorate_50r50w_1M.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
16 threadCount, READ, 74852, 41220, 41220, 41220, 0.3, 0.2, 0.4, 0.8, 2.7, 93.6, 1.8, 0.06069, 0, 1, 92, 92, 0, 1521
16 threadCount, WRITE, 25148, 13849, 13849, 13849, 0.3, 0.2, 0.4, 0.8, 3.3, 93.7, 1.8, 0.06069, 0, 1, 92, 92, 0, 1521
16 threadCount, total, 100000, 55069, 55069, 55069, 0.3, 0.2, 0.4, 0.8, 1.9, 93.7, 1.8, 0.06069, 0, 1, 92, 92, 0, 1521
24 threadCount, READ, 75161, 67239, 67239, 67239, 0.3, 0.2, 0.3, 0.6, 1.6, 50.3, 1.1, 0.12330, 0, 0, 0, 0, 0, 0
24 threadCount, WRITE, 24839, 22221, 22221, 22221, 0.2, 0.2, 0.3, 0.5, 1.9, 25.5, 1.1, 0.12330, 0, 0, 0, 0, 0, 0
24 threadCount, total, 100000, 89460, 89460, 89460, 0.3, 0.2, 0.3, 0.6, 1.8, 50.3, 1.1, 0.12330, 0, 0, 0, 0, 0, 0
36 threadCount, READ, 74515, 76617, 76617, 76617, 0.3, 0.2, 0.4, 0.6, 1.1, 174.0, 1.0, 0.00000, 0, 1, 173, 173, 0, 1538
36 threadCount, WRITE, 25485, 26204, 26204, 26204, 0.3, 0.2, 0.4, 0.7, 11.9, 174.0, 1.0, 0.00000, 0, 1, 173, 173, 0, 1538
36 threadCount, total, 100000, 102821, 102821, 102821, 0.3, 0.2, 0.4, 0.6, 1.0, 174.0, 1.0, 0.00000, 0, 1, 173, 173, 0, 1538
54 threadCount, READ, 75025, 92227, 92227, 92227, 0.4, 0.3, 0.5, 0.7, 25.2, 211.2, 0.8, 0.00000, 0, 1, 209, 209, 0, 1553
54 threadCount, WRITE, 24975, 30711, 30711, 30711, 0.4, 0.3, 0.5, 0.7, 25.4, 211.3, 0.8, 0.00000, 0, 1, 209, 209, 0, 1553
54 threadCount, total, 100000, 122928, 122928, 122928, 0.4, 0.3, 0.5, 0.7, 2.3, 211.3, 0.8, 0.00000, 0, 1, 209, 209, 0, 1553
81 threadCount, READ, 74565, 150728, 150728, 150728, 0.4, 0.3, 0.6, 0.9, 1.3, 25.7, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, WRITE, 25435, 50799, 50799, 50799, 0.4, 0.3, 0.6, 0.9, 2.0, 31.7, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
81 threadCount, total, 100000, 199721, 199721, 199721, 0.4, 0.3, 0.6, 0.9, 1.3, 31.7, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, READ, 74804, 162638, 162638, 162638, 0.5, 0.4, 0.8, 1.0, 1.2, 63.8, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, WRITE, 25196, 54783, 54783, 54783, 0.6, 0.4, 0.8, 1.3, 63.2, 63.8, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
121 threadCount, total, 100000, 217418, 217418, 217418, 0.6, 0.4, 0.8, 1.0, 1.5, 63.8, 0.5, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, READ, 74519, 206697, 206697, 206697, 0.6, 0.5, 1.1, 1.4, 23.4, 24.3, 0.4, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, WRITE, 25481, 70677, 70677, 70677, 0.7, 0.5, 1.1, 1.4, 23.3, 24.1, 0.4, 0.00000, 0, 0, 0, 0, 0, 0
181 threadCount, total, 100000, 277373, 277373, 277373, 0.7, 0.5, 1.1, 1.4, 23.3, 24.3, 0.4, 0.00000, 0, 0, 0, 0, 0, 0

CLUSTER47_blog_insert.log

1
2
3
4
5
6
7
8
9
10
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 232963, 7740, 7740, 10588, 0.4, 0.4, 0.6, 0.7, 0.9, 241.4, 30.1, 0.01115, 0, 3, 634, 634, 22, 4127
8 threadCount, insert, 436194, 14420, 14420, 19680, 0.5, 0.4, 0.6, 0.7, 1.0, 290.6, 30.2, 0.00780, 0, 7, 1677, 1677, 42, 9298
16 threadCount, insert, 720362, 23471, 23471, 32086, 0.6, 0.5, 0.8, 0.9, 1.9, 222.3, 30.7, 0.01669, 0, 14, 2259, 2259, 36, 19650
24 threadCount, insert, 762309, 24963, 24963, 34151, 0.8, 0.7, 1.1, 1.4, 6.3, 239.9, 30.5, 0.01428, 0, 16, 2494, 2494, 43, 22431
36 threadCount, insert, 782500, 25769, 25769, 35273, 1.3, 1.1, 2.0, 2.4, 13.0, 230.8, 30.4, 0.00666, 0, 15, 2454, 2454, 40, 20885
54 threadCount, insert, 779071, 25684, 25684, 35130, 2.0, 1.5, 3.6, 4.1, 13.7, 260.4, 30.3, 0.00705, 0, 18, 2607, 2607, 44, 30028
81 threadCount, insert, 799458, 25948, 25948, 35478, 3.0, 2.5, 5.4, 6.6, 144.4, 266.1, 30.8, 0.00924, 0, 15, 2181, 2181, 36, 20905
121 threadCount, insert, 806416, 26515, 26515, 36222, 4.4, 3.8, 8.5, 10.4, 100.2, 262.6, 30.4, 0.00663, 0, 15, 1936, 1936, 42, 20970
181 threadCount, insert, 813139, 26389, 26389, 36115, 6.7, 4.4, 15.7, 18.4, 166.1, 269.2, 30.8, 0.01099, 0, 15, 2067, 2067, 39, 20847

CLUSTER48_blog_mixed.log

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, insert, 37302, 1185, 1185, 1618, 0.5, 0.4, 0.7, 0.8, 1.4, 232.6, 31.5, 0.00958, 0, 9, 535, 535, 13, 16094
4 threadCount, singlepost, 73938, 2350, 2287, 2287, 0.8, 0.8, 1.1, 1.2, 1.4, 93.2, 31.5, 0.00958, 0, 9, 535, 535, 13, 16094
4 threadCount, timeline, 37531, 1193, 1161, 9236, 1.0, 0.9, 1.3, 1.4, 1.7, 93.3, 31.5, 0.00958, 0, 9, 535, 535, 13, 16094
4 threadCount, total, 148771, 4728, 4633, 13140, 0.8, 0.8, 1.1, 1.3, 1.4, 232.6, 31.5, 0.00958, 0, 9, 535, 535, 13, 16094
8 threadCount, insert, 74933, 2379, 2379, 3256, 0.5, 0.4, 0.7, 0.8, 7.1, 54.7, 31.5, 0.00967, 0, 13, 473, 473, 10, 20848
8 threadCount, singlepost, 151072, 4797, 4667, 4667, 0.8, 0.8, 1.0, 1.1, 7.6, 55.8, 31.5, 0.00967, 0, 13, 473, 473, 10, 20848
8 threadCount, timeline, 75451, 2396, 2331, 18748, 1.0, 0.9, 1.2, 1.4, 10.3, 55.8, 31.5, 0.00967, 0, 13, 473, 473, 10, 20848
8 threadCount, total, 301456, 9571, 9377, 26671, 0.8, 0.8, 1.1, 1.2, 6.1, 55.8, 31.5, 0.00967, 0, 13, 473, 473, 10, 20848
16 threadCount, insert, 145800, 4531, 4531, 6208, 0.5, 0.4, 0.7, 0.8, 1.2, 71.5, 32.2, 0.00600, 0, 21, 832, 832, 11, 34730
16 threadCount, singlepost, 294522, 9152, 8910, 8910, 0.8, 0.8, 1.1, 1.2, 6.7, 72.1, 32.2, 0.00600, 0, 21, 832, 832, 11, 34730
16 threadCount, timeline, 145130, 4510, 4390, 35766, 1.0, 1.0, 1.3, 1.4, 3.2, 72.4, 32.2, 0.00600, 0, 21, 832, 832, 11, 34730
16 threadCount, total, 585452, 18193, 17831, 50885, 0.8, 0.8, 1.1, 1.3, 6.2, 72.4, 32.2, 0.00600, 0, 21, 832, 832, 11, 34730
24 threadCount, insert, 205026, 6333, 6333, 8666, 0.5, 0.5, 0.8, 0.9, 2.5, 240.6, 32.4, 0.01018, 0, 23, 900, 900, 8, 37887
24 threadCount, singlepost, 407459, 12585, 12271, 12271, 0.9, 0.9, 1.1, 1.3, 2.5, 69.3, 32.4, 0.01018, 0, 23, 900, 900, 8, 37887
24 threadCount, timeline, 205264, 6340, 6182, 51336, 1.1, 1.0, 1.3, 1.5, 7.0, 69.7, 32.4, 0.01018, 0, 23, 900, 900, 8, 37887
24 threadCount, total, 817749, 25257, 24786, 72273, 0.9, 0.9, 1.2, 1.4, 5.4, 240.6, 32.4, 0.01018, 0, 23, 900, 900, 8, 37887
36 threadCount, insert, 288985, 8787, 8787, 12010, 0.6, 0.5, 0.9, 1.1, 7.9, 223.1, 32.9, 0.00892, 0, 32, 1210, 1210, 2, 50873
36 threadCount, singlepost, 575082, 17486, 17068, 17068, 1.0, 0.9, 1.2, 1.4, 9.0, 225.5, 32.9, 0.00892, 0, 32, 1210, 1210, 2, 50873
36 threadCount, timeline, 287981, 8756, 8548, 72597, 1.2, 1.1, 1.5, 1.7, 8.1, 226.4, 32.9, 0.00892, 0, 32, 1210, 1210, 2, 50873
36 threadCount, total, 1152048, 35029, 34403, 101674, 0.9, 0.9, 1.3, 1.5, 2.2, 226.4, 32.9, 0.00892, 0, 32, 1210, 1210, 2, 50873
54 threadCount, insert, 386007, 11602, 11602, 15868, 0.7, 0.6, 1.0, 1.3, 4.4, 50.5, 33.3, 0.00845, 0, 40, 1557, 1557, 2, 63491
54 threadCount, singlepost, 773934, 23262, 22762, 22762, 1.1, 1.0, 1.5, 1.8, 15.0, 51.3, 33.3, 0.00845, 0, 40, 1557, 1557, 2, 63491
54 threadCount, timeline, 387013, 11633, 11381, 99175, 1.3, 1.2, 1.7, 2.1, 18.7, 50.6, 33.3, 0.00845, 0, 40, 1557, 1557, 2, 63491
54 threadCount, total, 1546954, 46497, 45746, 137805, 1.1, 1.0, 1.5, 1.9, 13.0, 51.3, 33.3, 0.00845, 0, 40, 1557, 1557, 2, 63491
81 threadCount, insert, 440395, 12498, 12498, 17085, 1.2, 1.0, 1.5, 1.8, 15.2, 275.6, 35.2, 0.00697, 0, 43, 1707, 1707, 2, 68157
81 threadCount, singlepost, 881687, 25021, 24544, 24544, 1.6, 1.4, 2.0, 2.3, 37.1, 92.8, 35.2, 0.00697, 0, 43, 1707, 1707, 2, 68157
81 threadCount, timeline, 441963, 12542, 12299, 109925, 1.8, 1.6, 2.2, 2.6, 39.0, 92.8, 35.2, 0.00697, 0, 43, 1707, 1707, 2, 68157
81 threadCount, total, 1764045, 50060, 49340, 151554, 1.5, 1.4, 2.0, 2.4, 35.6, 275.6, 35.2, 0.00697, 0, 43, 1707, 1707, 2, 68157
121 threadCount, insert, 479643, 12574, 12574, 17174, 2.0, 1.8, 2.3, 2.8, 41.4, 277.5, 38.1, 0.00771, 0, 45, 1788, 1788, 1, 71366
121 threadCount, singlepost, 949679, 24896, 24466, 24466, 2.4, 2.2, 2.8, 3.3, 41.6, 239.1, 38.1, 0.00771, 0, 45, 1788, 1788, 1, 71366
121 threadCount, timeline, 476907, 12502, 12289, 112297, 2.6, 2.4, 3.0, 3.5, 42.9, 239.3, 38.1, 0.00771, 0, 45, 1788, 1788, 1, 71366
121 threadCount, total, 1906229, 49971, 49328, 153936, 2.3, 2.1, 2.8, 3.3, 41.8, 277.5, 38.1, 0.00771, 0, 45, 1788, 1788, 1, 71366
181 threadCount, insert, 534496, 12376, 12376, 16912, 3.2, 2.9, 3.5, 4.4, 45.5, 351.2, 43.2, 0.00806, 0, 54, 2139, 2164, 8, 84061
181 threadCount, singlepost, 1070775, 24793, 24419, 24419, 3.6, 3.3, 4.1, 5.4, 45.1, 112.1, 43.2, 0.00806, 0, 54, 2139, 2164, 8, 84061
181 threadCount, timeline, 535592, 12401, 12212, 113584, 3.8, 3.6, 4.3, 5.4, 42.7, 119.9, 43.2, 0.00806, 0, 54, 2139, 2164, 8, 84061
181 threadCount, total, 2140863, 49569, 49007, 154916, 3.5, 3.3, 4.1, 5.2, 44.2, 351.2, 43.2, 0.00806, 0, 54, 2139, 2164, 8, 84061
271 threadCount, insert, 621662, 12450, 12450, 17020, 4.9, 4.5, 5.3, 6.6, 49.1, 352.5, 49.9, 0.00709, 0, 65, 2670, 2670, 8, 105265
271 threadCount, singlepost, 1236932, 24773, 24448, 24448, 5.4, 5.0, 5.9, 7.4, 50.8, 200.2, 49.9, 0.00709, 0, 65, 2670, 2670, 8, 105265
271 threadCount, timeline, 615702, 12331, 12169, 114805, 5.7, 5.3, 6.2, 9.9, 57.4, 200.9, 49.9, 0.00709, 0, 65, 2670, 2670, 8, 105265
271 threadCount, total, 2474296, 49554, 49066, 156273, 5.4, 5.0, 5.9, 7.5, 52.9, 352.5, 49.9, 0.00709, 0, 65, 2670, 2670, 8, 105265

CLUSTER49_blog_q1.log

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, singlepost, 124438, 4137, 4088, 4088, 0.9, 0.9, 1.2, 1.3, 1.5, 36.0, 30.1, 0.00523, 0, 7, 235, 235, 18, 12655
8 threadCount, singlepost, 260639, 8663, 8560, 8560, 0.8, 0.9, 1.2, 1.3, 1.5, 31.4, 30.1, 0.00397, 0, 7, 173, 173, 2, 11432
16 threadCount, singlepost, 529703, 17591, 17380, 17380, 0.8, 0.8, 1.2, 1.3, 1.7, 27.4, 30.1, 0.00446, 0, 11, 251, 251, 1, 17981
24 threadCount, singlepost, 811028, 26917, 26599, 26599, 0.8, 0.7, 1.2, 1.4, 4.0, 27.0, 30.1, 0.00268, 0, 13, 280, 280, 2, 21261
36 threadCount, singlepost, 1140114, 37749, 37294, 37294, 0.9, 0.8, 1.3, 1.5, 3.9, 28.3, 30.2, 0.00807, 0, 16, 331, 331, 2, 26171
54 threadCount, singlepost, 1169908, 38732, 38262, 38262, 1.3, 1.2, 1.7, 1.9, 7.8, 25.9, 30.2, 0.00498, 0, 15, 300, 300, 2, 24535
81 threadCount, singlepost, 1180577, 39018, 38557, 38557, 2.0, 1.9, 2.3, 2.5, 18.2, 70.3, 30.3, 0.00604, 0, 16, 314, 314, 3, 26168
121 threadCount, singlepost, 1186818, 39141, 38676, 38676, 3.0, 3.0, 3.2, 3.4, 14.2, 32.9, 30.3, 0.00262, 0, 12, 223, 223, 4, 19624
181 threadCount, singlepost, 1196971, 39363, 38887, 38887, 4.5, 4.5, 4.7, 4.9, 17.0, 30.4, 30.4, 0.00113, 0, 13, 216, 216, 1, 21263
271 threadCount, singlepost, 1200807, 39256, 38777, 38777, 6.8, 6.7, 7.0, 7.3, 20.3, 169.8, 30.6, 0.00650, 0, 12, 200, 200, 1, 19628

CLUSTER410_blog_q2.log

1
2
3
4
5
6
7
8
9
10
11
             id, type,      total ops,    op/s,    pk/s,   row/s,    mean,     med,     .95,     .99,    .999,     max,   time,   stderr, errors,  gc: #,  max ms,  sum ms,  sdv ms,      mb
4 threadCount, timeline, 150606, 5005, 4945, 46957, 0.7, 0.7, 0.9, 1.0, 1.2, 18.1, 30.1, 0.00308, 0, 9, 134, 134, 1, 14732
8 threadCount, timeline, 303760, 10097, 9976, 94717, 0.7, 0.7, 0.9, 1.0, 1.2, 19.8, 30.1, 0.00180, 0, 12, 180, 180, 2, 19640
16 threadCount, timeline, 586886, 19491, 19260, 182833, 0.7, 0.7, 0.9, 1.0, 1.2, 23.1, 30.1, 0.00684, 0, 22, 292, 292, 2, 36025
24 threadCount, timeline, 842276, 27952, 27615, 262067, 0.8, 0.7, 1.0, 1.1, 3.5, 22.5, 30.1, 0.00723, 0, 32, 400, 400, 2, 52404
36 threadCount, timeline, 1159174, 38409, 37950, 360174, 0.8, 0.8, 1.1, 1.2, 5.6, 18.6, 30.2, 0.01793, 0, 41, 477, 477, 1, 67158
54 threadCount, timeline, 1212693, 40154, 39673, 376526, 1.3, 1.2, 1.5, 1.6, 8.3, 25.2, 30.2, 0.00919, 0, 42, 467, 475, 2, 67169
81 threadCount, timeline, 1216916, 40230, 39744, 377269, 1.9, 1.9, 2.1, 2.4, 11.4, 74.0, 30.2, 0.00547, 0, 42, 454, 454, 1, 68812
121 threadCount, timeline, 1228813, 40520, 40037, 379868, 2.9, 2.9, 3.1, 3.4, 10.5, 23.5, 30.3, 0.00332, 0, 42, 461, 461, 1, 68813
181 threadCount, timeline, 1236969, 40658, 40170, 381272, 4.4, 4.3, 4.6, 5.1, 13.7, 27.8, 30.4, 0.00334, 0, 42, 457, 457, 1, 68814
271 threadCount, timeline, 1231892, 40351, 39865, 378388, 6.6, 6.5, 6.8, 7.5, 29.2, 221.9, 30.5, 0.00637, 0, 42, 453, 453, 0, 68813

YCSB

https://github.com/cloudius-systems/osv/wiki/Benchmarking-Cassandra-and-other-NoSQL-databases-with-YCSB

基本命令:

1
2
加载数据:bin/ycsb load 数据库 -P workloada配置文件 -s
运行测试:bin/ycsb run 数据库 -P workloada配置文件 -s

Cassandra-1.0:

1
2
3
4
5
./bin/ycsb load cassandra-10 -p hosts="192.168.122.89" -P workloads/workloada -s > workloada_load_res.txt
./bin/ycsb run cassandra-10 -threads 10 -p hosts="192.168.122.89" -P workloads/workloada -s > workloada_run_res.txt

./bin/ycsb load cassandra-10 -P workloads/workloada -p hosts=localhost -p columnfamily=data > ./my-results/load-cassandra-a
./bin/ycsb run cassandra-10 -P workloads/workloada -p hosts=localhost -p columnfamily=data > ./my-results/run-cassandra-a

workloada配置文件示例:

1
2
3
4
5
6
7
8
9
10
11
hosts=192.168.6.53
cassandra.keyspace=ycsb
recordcount=100000
operationcount=100000
workload=com.yahoo.ycsb.workloads.CoreWorkload
readallfields=true
readproportion=0.5
updateproportion=0.5
scanproportion=0
insertproportion=0
requestdistribution=zipfian

Cassandra-2.0:

1
2
./bin/ycsb load cassandra2-cql -P workloada -s
./bin/ycsb run cassandra2-cql -P workloada -s

文章目录
  1. 1. cassandra-stress
    1. 1.1. 介绍
    2. 1.2. 单节点测试
      1. 1.2.1. 写入测试
      2. 1.2.2. 采用cql3协议写入
      3. 1.2.3. 读取测试
      4. 1.2.4. 没有预热读取
      5. 1.2.5. 给定一段时间测试最大可以读取多少条记录
      6. 1.2.6. 混合模式读大于写
      7. 1.2.7. 混合模式写大于读
    3. 1.3. 自定义schema
      1. 1.3.1. 示例
      2. 1.3.2. 只有写入
      3. 1.3.3. 读写混合模式
      4. 1.3.4. 只有读取
    4. 1.4. G1测试
    5. 1.5. 集群测试
  2. 2. YCSB