incubator-hama-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alejandro Claudio Esquivel Gonzalez (Commented) (JIRA)" <>
Subject [jira] [Commented] (HAMA-524) [GSoC 2012] Sparse Matrix-Vector multiplication (SpMV) on Hama
Date Wed, 11 Apr 2012 02:58:17 GMT


Alejandro Claudio Esquivel Gonzalez commented on HAMA-524:

So, you say that the task has been already assigned and there will not be change on that?
or that if you find a good idea to break this task into pieces you could accept more students?

There are different formats to process the Sparse Matrix-vector multiplication if the Sparse
matrix is in diagonal, tridiagonal form, band matrix, etc. I think you could identify the
best format for each matrix and split the conversion and operation of different formats into
different tasks. 

Also, there are more matrix operations that need different storage structures to be processed
efficiently, for example editing a single value of the matrix works better if it's in Yale

> [GSoC 2012] Sparse Matrix-Vector multiplication (SpMV) on Hama
> --------------------------------------------------------------
>                 Key: HAMA-524
>                 URL:
>             Project: Hama
>          Issue Type: New Feature
>          Components: bsp, examples, matrix
>            Reporter: Edward J. Yoon
>            Assignee: Mikalai Parafeniuk
>              Labels: gsoc, gsoc2012, newbie
> Implement Efficient and Fast SpMV algorithm which can be widely used in scientiļ¬c computing,
financial modeling, information retrieval, and others, using Hama Bulk Synchronous Parallel

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:!default.jspa
For more information on JIRA, see:


View raw message