zookeeper-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "maoling (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ZOOKEEPER-3264) Add a benchmark tool for zookeeper
Date Mon, 11 Mar 2019 11:26:00 GMT

    [ https://issues.apache.org/jira/browse/ZOOKEEPER-3264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16789434#comment-16789434
] 

maoling commented on ZOOKEEPER-3264:
------------------------------------

Look at an example from *redis*:

./redis-benchmark -h 127.0.0.1 -p 6379 -c 100 -n 1000
====== PING_INLINE ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

97.90% <= 1 milliseconds
100.00% <= 1 milliseconds
111111.12 requests per second

====== PING_BULK ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

97.10% <= 1 milliseconds
100.00% <= 1 milliseconds
124999.99 requests per second

====== SET ======
 1000 requests completed in 0.02 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

90.10% <= 9 milliseconds
91.10% <= 10 milliseconds
97.00% <= 11 milliseconds
100.00% <= 11 milliseconds
55555.56 requests per second

====== GET ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

100.00% <= 0 milliseconds
124999.99 requests per second

====== INCR ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

100.00% <= 0 milliseconds
124999.99 requests per second

====== LPUSH ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

90.70% <= 1 milliseconds
100.00% <= 1 milliseconds
124999.99 requests per second

====== RPUSH ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

100.00% <= 0 milliseconds
111111.12 requests per second

====== LPOP ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

99.80% <= 1 milliseconds
100.00% <= 1 milliseconds
124999.99 requests per second

====== RPOP ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

100.00% <= 0 milliseconds
124999.99 requests per second

====== SADD ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

60.10% <= 1 milliseconds
100.00% <= 1 milliseconds
76923.08 requests per second

====== HSET ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

98.90% <= 1 milliseconds
100.00% <= 1 milliseconds
124999.99 requests per second

====== SPOP ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

100.00% <= 0 milliseconds
142857.14 requests per second

====== LPUSH (needed to benchmark LRANGE) ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

99.50% <= 1 milliseconds
100.00% <= 1 milliseconds
124999.99 requests per second

====== LRANGE_100 (first 100 elements) ======
 1000 requests completed in 0.02 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

62.10% <= 1 milliseconds
94.50% <= 2 milliseconds
100.00% <= 3 milliseconds
52631.58 requests per second

====== LRANGE_300 (first 300 elements) ======
 1000 requests completed in 0.05 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

0.90% <= 1 milliseconds
10.00% <= 2 milliseconds
83.00% <= 3 milliseconds
92.00% <= 4 milliseconds
95.20% <= 5 milliseconds
98.30% <= 6 milliseconds
100.00% <= 6 milliseconds
20000.00 requests per second

====== LRANGE_500 (first 450 elements) ======
 1000 requests completed in 0.07 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

0.50% <= 1 milliseconds
1.10% <= 2 milliseconds
26.00% <= 3 milliseconds
83.80% <= 4 milliseconds
91.00% <= 5 milliseconds
95.70% <= 6 milliseconds
97.40% <= 7 milliseconds
98.70% <= 8 milliseconds
100.00% <= 8 milliseconds
14925.37 requests per second

====== LRANGE_600 (first 600 elements) ======
 1000 requests completed in 0.09 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

0.40% <= 1 milliseconds
0.50% <= 2 milliseconds
3.30% <= 3 milliseconds
22.40% <= 4 milliseconds
76.20% <= 5 milliseconds
85.10% <= 6 milliseconds
91.00% <= 7 milliseconds
94.60% <= 8 milliseconds
96.90% <= 9 milliseconds
98.30% <= 10 milliseconds
99.40% <= 11 milliseconds
100.00% <= 11 milliseconds
11235.96 requests per second

====== MSET (10 keys) ======
 1000 requests completed in 0.01 seconds
 100 parallel clients
 3 bytes payload
 keep alive: 1

63.30% <= 1 milliseconds
100.00% <= 1 milliseconds
83333.34 requests per second

> Add a benchmark tool for zookeeper
> ----------------------------------
>
>                 Key: ZOOKEEPER-3264
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-3264
>             Project: ZooKeeper
>          Issue Type: New Feature
>          Components: other
>            Reporter: maoling
>            Assignee: maoling
>            Priority: Major
>
> Reference:
> https://github.com/etcd-io/etcd/blob/master/tools/benchmark/cmd/range.go
> https://github.com/antirez/redis/blob/unstable/src/redis-benchmark.c
> https://github.com/phunt/zk-smoketest/blob/master/zk-latencies.py
> https://github.com/brownsys/zookeeper-benchmark/blob/master/src/main/java/edu/brown/cs/zkbenchmark/ZooKeeperBenchmark.java



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Mime
View raw message