Henggao’s HomePage
Henggao’s HomePage/
🏆
Experiment Cases/
🏆
Exper case(4)Cassandra测试
Search
🏆

Exper case(4)Cassandra测试

1. 数据测试50%read,50%update1. 1 参数设置1.2 参数说明2. YCSB测试2.1 load阶段2.2 run阶段2.3 结果2.3.1 100W单线程2.3.2 100W四线程2.3.2 100W16线程

1. 数据测试50%read,50%update

1. 1 参数设置

编辑测试文件workload_cassandra
测试100W条数据，单线程、4线程、16线程，修改threadcount =1，4，16
# YCSB参数
recordcount = 1000000
operationcount = 1000000
workload = site.ycsb.workloads.CoreWorkload
readproportion = 0.5
updateproportion = 0.5
insertproportion = 0
scanproportion = 0
requestdistribution = zipfian
threadcount = 4
readallfields = true

# Cassanra参数
hosts = 192.168.92.159,192.168.92.160,192.168.92.161
port = 9042
cassandra.keyspace = ycsb    
cassandra.readconsistencylevel = QUORUM
cassandra.writeconsistencylevel = QUORUM
cassandra.maxconnections = 100
cassandra.connecttimeoutmillis = 1000000000
cassandra.readtimeoutmillis  = 1000000000
  • Tips📢📢📢
    • hosts = 192.168.92.159,192.168.92.160,192.168.92.161后面不要有空格，有会报错
ycsb参数
github.com
https://github.com/baidu/tera/blob/master/doc/tools/ycsb.md
cassandra参数
github.com
https://github.com/brianfrankcooper/YCSB/tree/master/cassandra

1.2 参数说明

# Yahoo! Cloud System Benchmark
# Workload A: Update heavy workload
#   Application example: Session store recording recent actions
#   Read/update ratio: 50/50
#   Default data size: 1 KB records (10 fields, 100 bytes each, plus key)
#   Request distribution: zipfian
recordcount = 1000（表示load和run操作中，使用的YCSB实例记录数）
operationcount = 1000（表示load和run操作中，使用的YCSB实例操作数）
workload = com.yahoo.ycsb.workloads.CoreWorkload（要使用的工作负载类）
readproportion = 0.5（默认是0.95，表示的是进行read的操作占所有操作的比例）
updateproportion = 0.5（默认是0.05，表示的是进行update的操作占所有操作的比例）
insertproportion = 0（默认是0，表示的是进行insert的操作占所有操作的比例）
scanproportion = 0（默认是0，表示的是进行scan的操作占所有操作的比例）
requestdistribution = zipfian（默认是uniform，应该使用什么分布来选择要操作的记录：uniform均匀, zipfian, hotspot, sequential, exponential 和 latest）
threadcount = 2（默认值是1，表示YCSB客户端线程数）
readallfields = true（默认值是1，应该读取读取所有字段（true），只读取一个（false））

2. YCSB测试

2.1 load阶段

bin/ycsb.sh load cassandra-cql -P ./data_record/workloada_cassandra -p columnfamily=test_table -s > ./data_record/load_16threads.txt

2.2 run阶段

bin/ycsb.sh run cassandra-cql -P ./data_record/workloada_cassandra -p columnfamily=test_table -s > ./data_record/run_16threads.txt

2.3 结果

2.3.1 100W单线程

  • load大概33分钟
  • run大概33分钟
[OVERALL], RunTime(ms), 2003402
[OVERALL], Throughput(ops/sec), 499.15094424384125
[TOTAL_GCS_G1_Young_Generation], Count, 2410
[TOTAL_GC_TIME_G1_Young_Generation], Time(ms), 2990
[TOTAL_GC_TIME_%_G1_Young_Generation], Time(%), 0.14924613232890852
[TOTAL_GCS_G1_Old_Generation], Count, 0
[TOTAL_GC_TIME_G1_Old_Generation], Time(ms), 0
[TOTAL_GC_TIME_%_G1_Old_Generation], Time(%), 0.0
[TOTAL_GCs], Count, 2410
[TOTAL_GC_TIME], Time(ms), 2990
[TOTAL_GC_TIME_%], Time(%), 0.14924613232890852
[READ], Operations, 500720
[READ], AverageLatency(us), 2148.0499940086274
[READ], MinLatency(us), 839
[READ], MaxLatency(us), 78975
[READ], 95thPercentileLatency(us), 3361
[READ], 99thPercentileLatency(us), 4499
[READ], Return=OK, 500720
[CLEANUP], Operations, 1
[CLEANUP], AverageLatency(us), 2237440.0
[CLEANUP], MinLatency(us), 2236416
[CLEANUP], MaxLatency(us), 2238463
[CLEANUP], 95thPercentileLatency(us), 2238463
[CLEANUP], 99thPercentileLatency(us), 2238463
[UPDATE], Operations, 499280
[UPDATE], AverageLatency(us), 1843.2053276718475
[UPDATE], MinLatency(us), 718
[UPDATE], MaxLatency(us), 36319
[UPDATE], 95thPercentileLatency(us), 2869
[UPDATE], 99thPercentileLatency(us), 3913
[UPDATE], Return=OK, 499280

2.3.2 100W四线程

  • load大概10分钟
  • run大概10分钟
[OVERALL], RunTime(ms), 589489
[OVERALL], Throughput(ops/sec), 1696.384495724263
[TOTAL_GCS_G1_Young_Generation], Count, 725
[TOTAL_GC_TIME_G1_Young_Generation], Time(ms), 1204
[TOTAL_GC_TIME_%_G1_Young_Generation], Time(%), 0.20424469328520126
[TOTAL_GCS_G1_Old_Generation], Count, 0
[TOTAL_GC_TIME_G1_Old_Generation], Time(ms), 0
[TOTAL_GC_TIME_%_G1_Old_Generation], Time(%), 0.0
[TOTAL_GCs], Count, 725
[TOTAL_GC_TIME], Time(ms), 1204
[TOTAL_GC_TIME_%], Time(%), 0.20424469328520126
[READ], Operations, 500397
[READ], AverageLatency(us), 2614.141987262114
[READ], MinLatency(us), 744
[READ], MaxLatency(us), 184959
[READ], 95thPercentileLatency(us), 4727
[READ], 99thPercentileLatency(us), 6931
[READ], Return=OK, 500397
[CLEANUP], Operations, 4
[CLEANUP], AverageLatency(us), 564564.25
[CLEANUP], MinLatency(us), 3
[CLEANUP], MaxLatency(us), 2258943
[CLEANUP], 95thPercentileLatency(us), 2258943
[CLEANUP], 99thPercentileLatency(us), 2258943
[UPDATE], Operations, 499603
[UPDATE], AverageLatency(us), 2066.139596839891
[UPDATE], MinLatency(us), 625
[UPDATE], MaxLatency(us), 197759
[UPDATE], 95thPercentileLatency(us), 3727
[UPDATE], 99thPercentileLatency(us), 5559
[UPDATE], Return=OK, 499603
notion image

2.3.2 100W16线程

  • load大概2分钟
  • run大概2分钟
[OVERALL], RunTime(ms), 173600
[OVERALL], Throughput(ops/sec), 5760.3686635944705
[TOTAL_GCS_G1_Young_Generation], Count, 348
[TOTAL_GC_TIME_G1_Young_Generation], Time(ms), 982
[TOTAL_GC_TIME_%_G1_Young_Generation], Time(%), 0.565668202764977
[TOTAL_GCS_G1_Old_Generation], Count, 0
[TOTAL_GC_TIME_G1_Old_Generation], Time(ms), 0
[TOTAL_GC_TIME_%_G1_Old_Generation], Time(%), 0.0
[TOTAL_GCs], Count, 348
[TOTAL_GC_TIME], Time(ms), 982
[TOTAL_GC_TIME_%], Time(%), 0.565668202764977
[READ], Operations, 499535
[READ], AverageLatency(us), 2967.8904280981315
[READ], MinLatency(us), 765
[READ], MaxLatency(us), 1328127
[READ], 95thPercentileLatency(us), 5447
[READ], 99thPercentileLatency(us), 10103
[READ], Return=OK, 499535
[CLEANUP], Operations, 16
[CLEANUP], AverageLatency(us), 142424.1875
[CLEANUP], MinLatency(us), 2
[CLEANUP], MaxLatency(us), 2279423
[CLEANUP], 95thPercentileLatency(us), 344
[CLEANUP], 99thPercentileLatency(us), 2279423
[UPDATE], Operations, 500465
[UPDATE], AverageLatency(us), 2425.5168033728633
[UPDATE], MinLatency(us), 574
[UPDATE], MaxLatency(us), 779263
[UPDATE], 95thPercentileLatency(us), 4327
[UPDATE], 99thPercentileLatency(us), 8439
[UPDATE], Return=OK, 500465
 
Copyright 2022 Henggao Geng