[jira] [Commented] (DRILL-6301) Parquet Performance Analysis

["salim achouche (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/DRILL-6301?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16438061#comment-16438061 ] 

salim achouche commented on DRILL-6301:
---------------------------------------

Created the following [document|https://docs.google.com/document/d/1BSNem_ItP-Vxlr6auSP_iwwOLM9rwWZYxGwCsXi-IE8/edit?usp=sharing] which contains several JMH benchmarks to help answer the questions posed by this Jira.

> Parquet Performance Analysis
> ----------------------------
>
>                 Key: DRILL-6301
>                 URL: https://issues.apache.org/jira/browse/DRILL-6301
>             Project: Apache Drill
>          Issue Type: Task
>          Components: Storage - Parquet
>            Reporter: salim achouche
>            Assignee: salim achouche
>            Priority: Major
>             Fix For: 1.14.0
>
>
> _*Description -*_
>  * DRILL-5846 is meant to improve the Flat Parquet reader performance
>  * The associated implementation resulted in a 2x - 4x performance improvement
>  * Though during the review process ([pull request|[https://github.com/apache/drill/pull/1060])] few key questions arised
>  
> *_Intermediary Processing via Direct Memory vs Byte Arrays_*
>  * The main reasons for using byte arrays for intermediary processing is to a) avoid the high cost of the DrillBuf checks (especially the reference counting) and b) benefit from some observed Java optimizations when accessing byte arrays
>  * Starting with version 1.12.0, the DrillBuf enablement checks have been refined so that memory access and reference counting checks can be enabled independently
>  * Benchmarking of Java's Direct Memory unsafe method using JMH indicates the performance gap between heap vs direct memory  is very narrow except for few use-cases
>  * There are also concerns that the extra copy step (from direct memory into byte arrays) will have a negative effect on performance; note that this overhead was not observed using Intel's Vtune as the intermediary buffer were a) pinned to a single CPU, b) reused, and c) small enough to remain in the L1 cache during columnar processing.
> _*Goal*_ 
>  * The Flat Parquet reader is amongst the few Drill columnar operators
>  * It is imperative that we agree on the most optimal processing pattern so that the decisions that we take within this Jira are not only applied to Parquet but to all Drill columnar operators   
> _*Methodology*_ 
>  # Assess the performance impact of using intermediary byte arrays (as described above)
>  # Prototype a solution using Direct Memory and DrillBuf checks off, access checks on, all checks on
>  # Make an educated decision on which processing pattern should be adopted
>  # Decide whether it is ok to use Java's unsafe API (and through what mechanism) on byte arrays (when the use of byte arrays is a necessity)
>  
>  
>  
>  



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)