spark-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wes McKinney (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SPARK-13391) Use Apache Arrow as In-memory columnar store implementation
Date Sat, 20 Feb 2016 01:27:18 GMT

    [ https://issues.apache.org/jira/browse/SPARK-13391?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15155268#comment-15155268
] 

Wes McKinney commented on SPARK-13391:
--------------------------------------

Indeed, one of the major motivations of Arrow (for Python and R) is higher data throughput
between native pandas / R data-frame memory representation and Spark. I will be looking to
add C-level data marshaling algorithms between Arrow and pandas (via NumPy arrays) to the
Arrow codebase within the next couple of months. Will cross-post JIRAs as they develop

> Use Apache Arrow as In-memory columnar store implementation
> -----------------------------------------------------------
>
>                 Key: SPARK-13391
>                 URL: https://issues.apache.org/jira/browse/SPARK-13391
>             Project: Spark
>          Issue Type: New Feature
>          Components: Spark Core, SQL
>    Affects Versions: 1.6.0, 2.0.0
>            Reporter: Maciej BryƄski
>
> Idea.
> Apache Arrow (http://arrow.apache.org/) is Open Source implementation of inmemory columnar
store. It has APIs in many programming languages.
> We can think about using it in Apache Spark to avoid data (de-)serialization  
> when running PySpark (and R) UDFs.



--
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