[jira] [Updated] (SPARK-4129) Performance tuning in MultivariateOnlineSummarizer

["Xiangrui Meng (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/SPARK-4129?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Xiangrui Meng updated SPARK-4129:
---------------------------------
    Assignee: DB Tsai

> Performance tuning in MultivariateOnlineSummarizer
> --------------------------------------------------
>
>                 Key: SPARK-4129
>                 URL: https://issues.apache.org/jira/browse/SPARK-4129
>             Project: Spark
>          Issue Type: Improvement
>          Components: MLlib
>            Reporter: DB Tsai
>            Assignee: DB Tsai
>             Fix For: 1.2.0
>
>
> In MultivariateOnlineSummarizer, breeze's activeIterator is used to loop through the nonZero elements in the vector. However, activeIterator doesn't perform well due to lots of overhead. In this PR, native while loop is used for both DenseVector and SparseVector.
> The benchmark result with 20 executors using mnist8m dataset:
> Before:
> DenseVector: 48.2 seconds
> SparseVector: 16.3 seconds
> After:
> DenseVector: 17.8 seconds
> SparseVector: 11.2 seconds
> Since MultivariateOnlineSummarizer is used in several places, the overall performance gain in mllib library will be significant with this PR. 



--
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