spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sean Owen (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SPARK-18946) treeAggregate will be low effficiency when aggregate high dimension vectors in ML algorithm
Date Tue, 20 Dec 2016 13:15:58 GMT

    [ https://issues.apache.org/jira/browse/SPARK-18946?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15764191#comment-15764191
] 

Sean Owen commented on SPARK-18946:
-----------------------------------

I'm not sure what you're proposing as a fix though -- a big object is big, yes. It is already
compressed. Does it cause a problem that is deeper than that?

> treeAggregate will be low effficiency when aggregate high dimension vectors in ML algorithm
> -------------------------------------------------------------------------------------------
>
>                 Key: SPARK-18946
>                 URL: https://issues.apache.org/jira/browse/SPARK-18946
>             Project: Spark
>          Issue Type: Improvement
>          Components: ML, MLlib
>            Reporter: zunwen you
>              Labels: features
>
> In many machine learning algorithms, we have to treeAggregate large vectors/arrays due
to the large number of features. Unfortunately, the treeAggregate operation of RDD will be
low efficiency when the dimension of vectors/arrays is bigger than million. Because high dimension
of vector/array always occupy more than 100MB Memory, transferring a 100MB element among executors
is pretty low efficiency in Spark.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org


Mime
View raw message