incubator-hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Thomas Jungblut (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HAMA-221) A BSP matrix multiplication implementation.
Date Sat, 17 Sep 2011 12:27:08 GMT

    [ https://issues.apache.org/jira/browse/HAMA-221?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13107105#comment-13107105
] 

Thomas Jungblut commented on HAMA-221:
--------------------------------------

Played a bit with the algorithm, iterative and with the new fork/join framework in java7.

It is really cool, but it is just for quadratic matrices which length is a power of 2.
We should focus on a more general solution, shouldn't we?

I tagged the iterative solution with HAMA commentaries, how we can make a BSP out of it. 

https://github.com/thomasjungblut/thomasjungblut-common/blob/master/src/de/jungblut/math/matrix/StrassenMatrixMultiplication.java

If you are interested in the fork join implementation have a look at:

https://github.com/thomasjungblut/thomasjungblut-common/blob/master/src/de/jungblut/math/matrix/ForkJoinStrassenMatrixMultiplication.java

Maybe if I have a bit more time later I can write a BSP version.

> A BSP matrix multiplication implementation.
> -------------------------------------------
>
>                 Key: HAMA-221
>                 URL: https://issues.apache.org/jira/browse/HAMA-221
>             Project: Hama
>          Issue Type: Improvement
>          Components: bsp, matrix
>    Affects Versions: 0.3.0
>            Reporter: Edward J. Yoon
>
> Using the BSP model, we could improve the performance of mat-mat multiplication. Let's
evaluation it out on this issue.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message