[
https://issues.apache.org/jira/browse/SPARK-16321?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15383855#comment-15383855
]
Sameer Agarwal commented on SPARK-16321:
----------------------------------------
yes, The isNotNull(id) filter was automatically inferred from the existing
predicate: "id > 9000000". Can we try a different equivalent condition --
perhaps something along the lines of maxOf(id, 9000000) > 9000000 and see how
the 2 queries compare? This latter condition shouldn't generate the
isNotNull(id) filter and would help in isolating the cause of regression.
> Pyspark 2.0 performance drop vs pyspark 1.6
> -------------------------------------------
>
> Key: SPARK-16321
> URL: https://issues.apache.org/jira/browse/SPARK-16321
> Project: Spark
> Issue Type: Bug
> Components: PySpark
> Affects Versions: 2.0.0
> Reporter: Maciej Bryński
> Attachments: Spark16.nps, Spark2.nps, spark16._trace.png,
> spark2_nofilterpushdown.nps, spark2_trace.png, visualvm_spark16.png,
> visualvm_spark2.png, visualvm_spark2_G1GC.png
>
>
> I did some test on parquet file with many nested columns (about 30G in
> 400 partitions) and Spark 2.0 is 2x slower.
> {code}
> df = sqlctx.read.parquet(path)
> df.where('id > some_id').rdd.flatMap(lambda r: [r.id] if not r.id %100000
> else []).collect()
> {code}
> Spark 1.6 -> 2.3 min
> Spark 2.0 -> 4.6 min (2x slower)
> I used BasicProfiler for this task and cumulative time was:
> Spark 1.6 - 4300 sec
> Spark 2.0 - 5800 sec
> Should I expect such a drop in performance ?
> I don't know how to prepare sample data to show the problem.
> Any ideas ? Or public data with many nested columns ?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
