You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@systemml.apache.org by "Matthias Boehm (JIRA)" <ji...@apache.org> on 2017/03/21 19:41:41 UTC

[jira] [Closed] (SYSTEMML-1310) Parfor block partitioning (mini batches)

     [ https://issues.apache.org/jira/browse/SYSTEMML-1310?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Matthias Boehm closed SYSTEMML-1310.
------------------------------------

> Parfor block partitioning (mini batches)
> ----------------------------------------
>
>                 Key: SYSTEMML-1310
>                 URL: https://issues.apache.org/jira/browse/SYSTEMML-1310
>             Project: SystemML
>          Issue Type: Sub-task
>          Components: APIs, Runtime
>            Reporter: Matthias Boehm
>            Assignee: Matthias Boehm
>             Fix For: SystemML 1.0
>
>
> This task aims to add basic support for block partitioning of fixed size batches of rows or columns in the parfor optimizer and runtime. For now, we will only apply this for the spark execution backend. The basic supported form looks as follows, where we require that batches are aligned with the systemml blocksize.
> {code}
> N = 20;
> parfor(bi in 1 : nrow(X)/N) {
>    Xbi = X[((bi-1)*N+1):(bi*N), ];
>    R[bi, 1] = sum(Xbi);
> }
> {code}



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)