On Wed, Mar 14, 2012 at 7:53 AM, Bela Ban <[email protected]> wrote: > > > On 3/14/12 1:14 AM, Manik Surtani wrote: >> >> On 13 Mar 2012, at 03:28, Bela Ban wrote: >> >>> >>> >>> On 3/13/12 6:35 AM, Manik Surtani wrote: >>>> >>>> On 12 Mar 2012, at 08:03, Dan Berindei wrote: >>>> >>>>>> >>>>>> Well, probably not, because we only want to send keys to nodes that >>>>>> actually need to store them... >>>>>> >>>>> >>>>> Sending the whole tx as a multicast would certainly be more efficient >>>>> than what we do now with lots of targets. >>>>> With unicasts we could send only the minimum required data to each >>>>> target, but that computation would be complex and error-prone. >>>> >>>> Well, this is what ANYCAST was all about initially, where JGroups would >>>> decide, based on the recipient list versus the total cluster size, on >>>> whether to send multiple unicasts or a multicast. But we didn't end up >>>> doing this in the end, perhaps we need to revisit. >>>> >>>> Bela, I'm guessing this thread was prompted by the poor performance in >>>> DIST that was reported, right? I'd like to profile the test provided to >>>> understand where we should be looking in the first place. E.g., is it the >>>> fact that we have too many RPCs? Or maybe a locking/concurrency issue >>>> elsewhere, etc. If you have done any of this analysis already, we should >>>> talk about that. >>> >>> >>> Yes. My current findings are that we're doing unneeded sync RPCs even if >>> <async.../> is defined. I'll run this with the latest Infinispan and see >>> if it's still the case (Galder mentioned this was gone in 5.2 master). >>> Also, there should be performance improvements by locking only the >>> primary owners (changes by Mircea and/or Dan), so I'll need to re-run >>> with the latest and greatest of Infinispan (and JGroups). >> >> Any such changes should also be in 5.1.x. If you do see any sync RPCs >> (except perhaps remote GETs) when running in async mode, then that is almost >> certainly a bug. > > > Yes, this is a bug, I confirmed that with Dan yesterday. He's aware of > it and the fix is simple, so I expect this will be fixed shortly. The > workaround is to define <stateTransfer .../>. >
Indeed, somehow I missed this thread yesterday, but based on my chat with Bela I've created https://issues.jboss.org/browse/ISPN-1910 and https://github.com/infinispan/infinispan/pull/998. Cheers Dan _______________________________________________ infinispan-dev mailing list [email protected] https://lists.jboss.org/mailman/listinfo/infinispan-dev
