On 11/30/06, Michael Bauroth <[EMAIL PROTECTED]> wrote:
> There's no estimate so far actually. What we actually have to focus is > documentation of MINA 1.0. You won't see any difference in performance if > your connection last for a long time. It only affects when the number of > connections are small and disconnection occurs often. Of course, it's > entirely up to your team whether to use trunk or not. :) > > Trustin If you refer to my last post: There is a measurable difference between 0.x and newer versions, but non documented from my side between 1.x and trunk. The mentioned 10%-20% come here from a test with Java 6 against Java 5, which I have made with the trunk. The test performed very well under heavy load (24.000 connections sent every 10 seconds 300 Bytes of bundeled string messages, which were demultiplexed to different decoders, decoded to POJOs and written to a MySQL database as batches). All this happened on a simple Xeon Dualcore with at least under 50 percent of load.
This is impressive! I guess I missed your last post. Sorry for that. Probably we could share our performance test results to the CWiki. Let me create a dedicated page. You could post your own if you've faxed your CLA. Thanks, Trustin -- what we call human nature is actually human habit -- http://gleamynode.net/ -- PGP key fingerprints: * E167 E6AF E73A CBCE EE41 4A29 544D DE48 FE95 4E7E * B693 628E 6047 4F8F CFA4 455E 1C62 A7DC 0255 ECA6
