[
https://issues.apache.org/jira/browse/SPARK-6019?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14516026#comment-14516026
]
Xiangrui Meng commented on SPARK-6019:
--------------------------------------
The linking part is actually quite complicated. I think we should only provide
a link to netlib-java (https://github.com/fommil/netlib-java) for instructions.
> Document linking with high-performance BLAS libraries
> -----------------------------------------------------
>
> Key: SPARK-6019
> URL: https://issues.apache.org/jira/browse/SPARK-6019
> Project: Spark
> Issue Type: Documentation
> Components: Documentation, MLlib
> Affects Versions: 1.3.0
> Reporter: Joseph K. Bradley
>
> Even though we may not be able to provide high-performance BLAS libraries
> (due to licenses, system dependencies, etc.), we could provide better
> documentation about how to link with such libraries on various systems. This
> could be a new section in the MLlib programming guide.
> Tuned libraries can be much faster than default ones. See discussion [on
> this dev@spark thread |
> http://apache-spark-developers-list.1001551.n3.nabble.com/Using-CUDA-within-Spark-boosting-linear-algebra-td10481.html].
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
