db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dyre Tjeldvoll (JIRA)" <j...@apache.org>
Subject [jira] Commented: (DERBY-3092) Use java.util.concurrent in TransactionTable and XactFactory to improve scalability
Date Fri, 28 Sep 2007 13:24:50 GMT

    [ https://issues.apache.org/jira/browse/DERBY-3092?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12531010
] 

Dyre Tjeldvoll commented on DERBY-3092:
---------------------------------------

The test was performed on a SunFire T2000 with 8 cores and 4
hardware threads per core, using jvm version 1.6.0_02. The client
was run (in embedded mode) for 300 seconds with a 30 second
warmup, with 32 clients. The CPU utilization of the jvm process
was observed with top:

10.3.1.4:                     
CPU utilization: 35.3% 
Throughput: 37,512.6 TPS

With new buffer manager:
CPU utilization: 82.8%
Throughput: 132,444 TPS

With new buffer manager and java.util.concurrent in
TransactionTable and XactFfactory: 
CPU utilization: 97.4%
Throughput: 159.720 TPS

> Use java.util.concurrent in TransactionTable and XactFactory to improve scalability
> -----------------------------------------------------------------------------------
>
>                 Key: DERBY-3092
>                 URL: https://issues.apache.org/jira/browse/DERBY-3092
>             Project: Derby
>          Issue Type: Improvement
>          Components: Performance, Store
>    Affects Versions: 10.3.1.4
>            Reporter: Dyre Tjeldvoll
>
> Running scalability tests with the client and buffer manager from DERBY-2911 shows that
access to the TransactionTable.trans (a Hashtable) and XactFactory.tranId (a shared long)
are the next major sources of contention. 

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message