[
https://issues.apache.org/jira/browse/HBASE-25658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17300111#comment-17300111
]
Duo Zhang commented on HBASE-25658:
-----------------------------------
Setup a cluster with 10 nodes, run PE tool for write, scan and read. The
commands are
{code}
./bin/hbase pe --nomapred --presplit=10 randomWrite 10
./bin/hbase pe --nomapred scan 10
./bin/hbase pe --nomapred --rows=20000 randomRead 10
{code}
For the current master, the results are
{noformat}
Write: 196367 row per second
[ Avg TPS/QPS] 187356 row per second
[ Avg TPS/QPS] 189755 row per second
[ Avg TPS/QPS] 211989 row per second
Scan: 72239 row per second
[ Avg TPS/QPS] 76148 row per second
[ Avg TPS/QPS] 66757 row per second
[ Avg TPS/QPS] 73812 row per second
Read: 2656 row per second
[ Avg TPS/QPS] 2632 row per second
[ Avg TPS/QPS] 2650 row per second
[ Avg TPS/QPS] 2685 row per second
{noformat}
For HBASE-22120 branch with trace disabled
{noformat}
Write: 201112 row per second
[ Avg TPS/QPS] 218045 row per second
[ Avg TPS/QPS] 201130 row per second
[ Avg TPS/QPS] 184162 row per second
Scan: 68004 row per second
[ Avg TPS/QPS] 68977 row per second
[ Avg TPS/QPS] 65415 row per second
[ Avg TPS/QPS] 69619 row per second
Read: 2636 row per second
[ Avg TPS/QPS] 2620 row per second
[ Avg TPS/QPS] 2629 row per second
[ Avg TPS/QPS] 2660 row per second
{noformat}
> Basic benchmark to show the impact on performance for tracing
> -------------------------------------------------------------
>
> Key: HBASE-25658
> URL: https://issues.apache.org/jira/browse/HBASE-25658
> Project: HBase
> Issue Type: Sub-task
> Reporter: Duo Zhang
> Assignee: Duo Zhang
> Priority: Major
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
