You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@arrow.apache.org by "Antoine Pitrou (Jira)" <ji...@apache.org> on 2020/02/03 15:37:00 UTC

[jira] [Commented] (ARROW-7748) [C++] [Cuda] Cache CUDA contexts

    [ https://issues.apache.org/jira/browse/ARROW-7748?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17029041#comment-17029041 ] 

Antoine Pitrou commented on ARROW-7748:
---------------------------------------

This issue came while reviewing the PR for the Device API.

> [C++] [Cuda] Cache CUDA contexts
> --------------------------------
>
>                 Key: ARROW-7748
>                 URL: https://issues.apache.org/jira/browse/ARROW-7748
>             Project: Apache Arrow
>          Issue Type: Improvement
>          Components: C++, GPU
>            Reporter: Antoine Pitrou
>            Priority: Major
>
> CUDA contexts can be expensive to instantiate, maybe we should cache them (probably only the primary contexts).



--
This message was sent by Atlassian Jira
(v8.3.4#803005)