diff options
| author | Davies Liu <davies@databricks.com> | 2015-12-21 14:21:43 -0800 |
|---|---|---|
| committer | Davies Liu <davies.liu@gmail.com> | 2015-12-21 14:21:43 -0800 |
| commit | 29cecd4a42f6969613e5b2a40f2724f99e7eec01 (patch) | |
| tree | 261b40272c991649ce584b7cec2056dcd2b6cf1c /.rat-excludes | |
| parent | d655d37ddf59d7fb6db529324ac8044d53b2622a (diff) | |
| download | spark-29cecd4a42f6969613e5b2a40f2724f99e7eec01.tar.gz spark-29cecd4a42f6969613e5b2a40f2724f99e7eec01.tar.bz2 spark-29cecd4a42f6969613e5b2a40f2724f99e7eec01.zip | |
[SPARK-12388] change default compression to lz4
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
cc rxin
Author: Davies Liu <davies@databricks.com>
Closes #10342 from davies/lz4.
Diffstat (limited to '.rat-excludes')
| -rw-r--r-- | .rat-excludes | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/.rat-excludes b/.rat-excludes index 7262c960ed..3544c0fc3d 100644 --- a/.rat-excludes +++ b/.rat-excludes @@ -84,3 +84,4 @@ gen-java.* org.apache.spark.sql.sources.DataSourceRegister org.apache.spark.scheduler.SparkHistoryListenerFactory .*parquet +LZ4BlockInputStream.java |