| Summary: | Broker needs better coalescing of outbound messages. | ||
|---|---|---|---|
| Product: | Red Hat Enterprise MRG | Reporter: | Ken Giusti <kgiusti> |
| Component: | qpid-cpp | Assignee: | Ken Giusti <kgiusti> |
| Status: | NEW --- | QA Contact: | Zdenek Kraus <zkraus> |
| Severity: | low | Docs Contact: | |
| Priority: | medium | ||
| Version: | Development | CC: | iboverma, jross, mcressma, zkraus |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | qpid-cpp-0.18-1 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | Type: | --- | |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Bug Depends On: | |||
| Bug Blocks: | 794748, 808134 | ||
|
Description
Ken Giusti
2012-02-15 18:24:44 UTC
Results of performance measure by qpid-cpp-benchmark using qpid-send, qpid-receive on qpid-0.14-22 vs. qpid-0.18-13 shows: RHEL 5.9 i686 and x86_64 have 3~20%, but mostly 10% performance boost, and in all cases they have positive improvement on the other hand RHEL 6.3 i686 have about -5% performance loss in most cases RHEL 6.3 x86_64 have around 50:50 distribution of improvement and losses, thus I assume that the change has no real effect on performance the measurements on RHEL 6.3 i686 and x86_64 was measured/runned twice to make check, both runs did show same results. for RHEL6 this is not an improvement, moving back -> ASSIGNED We're going to address this in the next release. I've performed some additional performance measurement between upstream version of qpidd. Specifically between r1244503 and r1244504, that introduces the relevant change (https://reviews.apache.org/r/3897/diff/1/?file=74938#file74938line102). I we see no significant performance improvement nor loss. |