[
https://issues.apache.org/jira/browse/SPARK-22221?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16195736#comment-16195736
]
Li Jin edited comment on SPARK-22221 at 10/7/17 2:49 PM:
---------------------------------------------------------
-I think we should also add to the document is what are the behavior
difference of arrow vs non-arrow serialization (if any).- (This is in
description already)
Just as a reminder, in the current state, there are behavior difference with
array and struct type between arrow and non-arrow version.
Array:
{code:java}
non-Arrow:
In [47]: type(df2.toPandas().array[0])
Out[47]: list
Arrow:
In [45]: type(df2.toPandas().array[0])
Out[45]: numpy.ndarray
{code}
Struct:
{code:java}
non-Arrow:
In [35]: type(df.toPandas().struct[0])
Out[35]: pyspark.sql.types.Row
Arrow:
In [37]: type(df.toPandas().struct[0])
Out[37]: dict
{code}
was (Author: icexelloss):
-I think we should also add to the document is what are the behavior
difference of arrow vs non-arrow serialization (if any).- (This is in
description already)
In the current state, there are difference in array and struct type between
arrow and non-arrow version.
Array:
{code:java}
non-Arrow:
In [47]: type(df2.toPandas().array[0])
Out[47]: list
Arrow:
In [45]: type(df2.toPandas().array[0])
Out[45]: numpy.ndarray
{code}
Struct:
{code:java}
non-Arrow:
In [35]: type(df.toPandas().struct[0])
Out[35]: pyspark.sql.types.Row
Arrow:
In [37]: type(df.toPandas().struct[0])
Out[37]: dict
{code}
> Add User Documentation for Working with Arrow in Spark
> ------------------------------------------------------
>
> Key: SPARK-22221
> URL: https://issues.apache.org/jira/browse/SPARK-22221
> Project: Spark
> Issue Type: Sub-task
> Components: PySpark, SQL
> Affects Versions: 2.3.0
> Reporter: Bryan Cutler
>
> There needs to be user facing documentation that will show how to enable/use
> Arrow with Spark, what the user should expect, and describe any differences
> with similar existing functionality.
> A comment from Xiao Li on https://github.com/apache/spark/pull/18664
> Given the users/applications contain the Timestamp in their Dataset and their
> processing algorithms also need to have the codes based on the corresponding
> time-zone related assumptions.
> * For the new users/applications, they first enabled Arrow and later hit an
> Arrow bug? Can they simply turn off spark.sql.execution.arrow.enable? If not,
> what should they do?
> * For the existing users/applications, they want to utilize Arrow for better
> performance. Can they just turn on spark.sql.execution.arrow.enable? What
> should they do?
> Note Hopefully, the guides/solutions are user-friendly. That means, it must
> be very simple to understand for most users.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
