GitHub user omalley opened a pull request:
https://github.com/apache/orc/pull/189
ORC-210 Add new encodings and benchmarks for new double encoding.
This extends Teddy's pull request by adding:
* Extended write benchmark.
* Added read benchmark
* Added new encodings:
+ plainV2, which is the same data format but faster implementation
+ fpcV2, which is the standard FPC implementation
+ flip, which rotates the bytes in a 8x8 matrix
+ split, which creates rle streams for sign, exponent, and mantissa
* Added new datasets.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/omalley/orc orc-210
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/orc/pull/189.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #189
----
commit ca6021511d38a3db87ac3a6d5b086155c05abf70
Author: Teddy Choi <pudi...@gmail.com>
Date: 2017-08-22T17:27:35Z
ORC-210: Add encoding for Double, Float.
commit 7a93e7f54e978ac3efb369d538a869775468a1bc
Author: Owen O'Malley <omal...@apache.org>
Date: 2017-10-26T23:41:43Z
ORC-210. Adding new double encoding.
* Moved implementations out of core until we pick one.
* Extended write benchmark.
* Added read benchmark
* Added new encodings:
+ plainV2, which is the same data format but faster implementation
+ fpcV2, which is the standard FPC implementation
+ flip, which rotates the bytes in a 8x8 matrix
+ split, which creates rle streams for sign, exponent, and mantissa
* Added new datasets.
----
---
