[infinispan-dev] New bundler performance

[Pedro Ruivo]

Hi Radim,

On 06/13/2013 09:24 AM, Radim Vansa wrote:
> Seems that it's somewhat better (6050 writes/s) but not at the previous level.

during this run, have you used the old bundler in JGroups?

> I am trying to cherry-pick it directly on the commit that failed it and see where is the difference, then maybe I'll have to generate alternative branch with ISPN-3221 commit after ISPN-2848 73da108cdcf9db4f3edbcd6dbda6938d6e45d148 and iterate to next regression origin.
>
> Pedro: I've noticed that in JGroupsTransport:577 you're changing the third parameter of CommandAwareRpcDispatcher.constructMessage from false (as found in 113842c8cf91cbb5a1bbd26e05fab7024fdec081) to true (as appears in later commits), is this desired?

It was not changed by me but yes I think that is ok. See ISPN-3043 
(https://issues.jboss.org/browse/ISPN-3043)

Cheers,
Pedro

ps. it's holiday in Lisbon so I'll not be online all day.

>
> Radim
>
> ----- Original Message -----
> | From: "Mircea Markus" <mmarkus at redhat.com>
> | To: "infinispan -Dev List" <infinispan-dev at lists.jboss.org>
> | Sent: Wednesday, June 12, 2013 5:51:13 PM
> | Subject: Re: [infinispan-dev] New bundler performance
> |
> | Radim, can you please give Pedro's branch a try?
> | https://github.com/infinispan/infinispan/pull/1896
> | If all good we'll release CR2 from that one.
> |
> | On 12 Jun 2013, at 16:02, Bela Ban <bban at redhat.com> wrote:
> | >
> | >
> | > On 06/12/2013 10:57 AM, Mircea Markus wrote:
> | >
> | >>> I think we should treat this as a bug and I'll look into this, but I
> | >>> won't have much time until after JBW. If you want to write code that can
> | >>> handle both old and new bundler, that would be good, although that code
> | >>> should be removed after the bug has been fixed.
> | >> The only thing I'm aware of that's needed to support old bundling is to
> | >> add DONT_BUNDLE to sync messages, which seems pretty straight forward.
> | >> I remember discussing that DONT_BUNDLE + sync messages + new bundler is
> | >> suboptimal for some reason (batching?) - am I wrong saying that?
> | >
> | >
> | > This will generate more messages than necessary if multiple threads are
> | > invoking sync RPCs, but other than that it should be ok. The main
> | > advantage of the new bundler is that we can remove the need for
> | > DONT_BUNDLE.
> | >
> | >
> | >> We're currently holding the 5.3.0.CR2 release for this. I'm thinking to:
> | >> - support old bundling as well
> | >> - based on the when the jgroups perf problem is fixed we might release CR3
> | >> or include the fix in 5.3.1.Final
> | >
> | > OK, sounds good
> | >
> | > --
> | > Bela Ban
> | > Lead JGroups / Clustering Team
> | > JBoss
> | > _______________________________________________
> | > infinispan-dev mailing list
> | > infinispan-dev at lists.jboss.org
> | > https://lists.jboss.org/mailman/listinfo/infinispan-dev
> |
> | Cheers,
> | --
> | Mircea Markus
> | Infinispan lead (www.infinispan.org)
> |
> |
> |
> |
> |
> | _______________________________________________
> | infinispan-dev mailing list
> | infinispan-dev at lists.jboss.org
> | https://lists.jboss.org/mailman/listinfo/infinispan-dev
> |
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev
>