cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ariel Weisberg (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-8692) Coalesce intra-cluster network messages
Date Fri, 30 Jan 2015 19:08:35 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-8692?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14299062#comment-14299062
] 

Ariel Weisberg edited comment on CASSANDRA-8692 at 1/30/15 7:07 PM:
--------------------------------------------------------------------

I've asked the two people who have tried the patch to post their experience. I personally
don't have a passion for which versions it goes in, but I am willing to do whatever work is
required to get it 2.1 ready.

I think it's better for stakeholders to decide the risk/reward for 2.1 that is acceptable
to them. It's hard to for a non-stakeholder like me to advocate for it. The one thing I will
say is that there is performance and then there is doubling performance. Granted that is only
in some configurations.

My benchmarks are in CASSANDRA-8457 and here are some of them.

https://issues.apache.org/jira/browse/CASSANDRA-8457?focusedCommentId=14266402&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14266402
In AWS without enhanced networking
||Coalesce window microseconds|Throughput||
|250| 502614|
|200| 496206|
|150| 487195|
|100| 423415|
|50| 326648|
|25| 308175|
|12| 292894|
|6| 268456|
|0| 153688|

https://issues.apache.org/jira/browse/CASSANDRA-8457?focusedCommentId=14266963&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14266963
In GCE
||Coalesce window microseconds| Throughput||
|200| 149791|
|150| 148755|
|100| 147678|
|50| 142565|
|25| 144542|
|12| 141679|
|6| 142626|
|0| 133905|
|0| 132849|


was (Author: aweisberg):
I've asked the two people who have tried the patch to post their experience. I personally
don't have a passion for which versions it goes in, but I am willing to whatever work is required
to get it 2.1 ready. I think it's better for stakeholders to decide the risk/reward for 2.1
that is acceptable to them. It's hard to for a non-stakeholder like me to advocate for it.
The one thing I will say is that there is performance and then there is doubling performance.
Granted in only some configurations.

My benchmarks are in CASSANDRA-8457 and here are some of them.

https://issues.apache.org/jira/browse/CASSANDRA-8457?focusedCommentId=14266402&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14266402
In AWS without enhanced networking
||Coalesce window microseconds|Throughput||
|250| 502614|
|200| 496206|
|150| 487195|
|100| 423415|
|50| 326648|
|25| 308175|
|12| 292894|
|6| 268456|
|0| 153688|

https://issues.apache.org/jira/browse/CASSANDRA-8457?focusedCommentId=14266963&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14266963
In GCE
||Coalesce window microseconds| Throughput||
|200| 149791|
|150| 148755|
|100| 147678|
|50| 142565|
|25| 144542|
|12| 141679|
|6| 142626|
|0| 133905|
|0| 132849|

> Coalesce intra-cluster network messages
> ---------------------------------------
>
>                 Key: CASSANDRA-8692
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-8692
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Ariel Weisberg
>            Assignee: Ariel Weisberg
>
> While researching CASSANDRA-8457 we found that it is effective and can be done without
introducing additional latency at low concurrency/throughput.
> The patch from that was used and found to be useful in a real life scenario so I propose
we implement this in 2.1 in addition to 3.0.
> The change set is a single file and is small enough to be reviewable.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message