[ 
https://issues.apache.org/jira/browse/IGNITE-13337?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17185892#comment-17185892
 ] 

Aleksey Plekhanov commented on IGNITE-13337:
--------------------------------------------

[~Artukhov], can you please run benchmarks for the following commits:
99b0e0143e0 (IGNITE-13060 Tracing: initial implementation) 
65c30ec6947 (IGNITE-12568 MessageFactory is refactored in order to detect 
registration of message with the same direct type)
We have performance drop about 2-3% for each of these commits, but measurement 
error is high, so we don't completely sure.

> Performance drop in ignite-2.9 vs Ignite 2.8.1
> ----------------------------------------------
>
>                 Key: IGNITE-13337
>                 URL: https://issues.apache.org/jira/browse/IGNITE-13337
>             Project: Ignite
>          Issue Type: Bug
>          Components: cache, sql
>    Affects Versions: 2.9
>         Environment: 9 equal physical hosts (2x Xeon E5-2609 v4, 96Gb RAM, 
> 3x800Gb SSD RAID0, 10Gb/s ethernet)
> 3 server nodes, one node per physical host
> 6 client nodes, one node per physical host, 64 load threads
>            Reporter: Ivan Artukhov
>            Priority: Critical
>         Attachments: caches.xml, driver.xml, in-mem-FULL_SYNC.png, 
> in-mem-PRIMARY_SYNC.png, pds-LOG_ONLY-FULL_SYNC.png, 
> pds-LOG_ONLY-PRIMARY_SYNC.png, server.xml
>
>
> Yardstick benchmarks show up to 7% performance drop in branch {{ignite-2.9}} 
> (rev. 6592dfa5) vs Ignite 2.8.1. Please see the attached screenshots.
> The benchmarks in use:
> {noformat}
> IgnitePutBenchmark
> IgniteGetBenchmark
> IgnitePutTxImplicitBenchmark
> IgnitePutGetTxBenchmark -optimistic
> IgnitePutGetTxBenchmark -pessimistic
> IgnitePutGetTxBenchmark -optimistic -serializable
> IgniteSqlQueryBenchmark
> IgnitePutAllTxBenchmark
> IgnitePutAllBenchmark
> {noformat}



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to