[
https://issues.apache.org/jira/browse/SPARK-12388?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Davies Liu updated SPARK-12388:
-------------------------------
Description:
According the benchmark [1], LZ4-java could be 80% (or 30%) faster than Snappy.
After changing the compressor to LZ4, I saw 20% improvement on end-to-end time
for a TPCDS query (Q4).
[1] https://github.com/ning/jvm-compressor-benchmark/wiki
was:
According the benchmark [1], LZ4-java could be 80% (or 30%) faster than Snappy.
After changing the compressor to LZ4, I saw 25% improvement on end-to-end time
for a TPCDS query (Q4).
[1] https://github.com/ning/jvm-compressor-benchmark/wiki
> Change default compressor to LZ4
> --------------------------------
>
> Key: SPARK-12388
> URL: https://issues.apache.org/jira/browse/SPARK-12388
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Davies Liu
>
> According the benchmark [1], LZ4-java could be 80% (or 30%) faster than
> Snappy.
> After changing the compressor to LZ4, I saw 20% improvement on end-to-end
> time for a TPCDS query (Q4).
> [1] https://github.com/ning/jvm-compressor-benchmark/wiki
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
