You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hama.apache.org by "Edward J. Yoon" <ed...@apache.org> on 2011/06/24 09:23:51 UTC
Bench test result.
Hi,
Here's my simple test result.
* Same cluster condition: 16 physical nodes, 256 cores, 10G network
* Bench example arguments: (16 bytes data, 1000 messages, 1024 supersteps)
Hama 0.2-incubating: 979.198 seconds
Hama 0.3-incubating: 874.016 seconds
Miklos's message-bundle patch seems very effective.
Thanks.
--
Best Regards, Edward J. Yoon
@eddieyoon