diff options
| author | Wenchen Fan <wenchen@databricks.com> | 2016-01-17 09:11:43 -0800 |
|---|---|---|
| committer | Davies Liu <davies.liu@gmail.com> | 2016-01-17 09:11:43 -0800 |
| commit | cede7b2a1134a6c93aff20ed5625054d988d3659 (patch) | |
| tree | 2364f0f142eaf0cf559ce21da6ead9a3cd2bb31e /R/run-tests.sh | |
| parent | 3c0d2365d57fc49ac9bf0d7cc9bd2ef633fb5fb6 (diff) | |
| download | spark-cede7b2a1134a6c93aff20ed5625054d988d3659.tar.gz spark-cede7b2a1134a6c93aff20ed5625054d988d3659.tar.bz2 spark-cede7b2a1134a6c93aff20ed5625054d988d3659.zip | |
[SPARK-12860] [SQL] speed up safe projection for primitive types
The idea is simple, use `SpecificMutableRow` instead of `GenericMutableRow` as result row for safe projection.
A simple benchmark shows about 1.5x speed up for primitive types, code: https://gist.github.com/cloud-fan/fa77713ccebf0823b2ab#file-safeprojectionbenchmark-scala
Author: Wenchen Fan <wenchen@databricks.com>
Closes #10790 from cloud-fan/safe-projection.
Diffstat (limited to 'R/run-tests.sh')
0 files changed, 0 insertions, 0 deletions