[jira] [Commented] (HIVE-10179) Optimization for SIMD instructions in Hive

["Teddy Choi (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HIVE-10179?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14681508#comment-14681508 ] 

Teddy Choi commented on HIVE-10179:
-----------------------------------

Hello [~chengxiang li].

I found there is more room for SIMD optimization. Followings are just some examples.

|| File || Before || After ||
| NotCol.java | {{~vector\[i\] & 1}} | {{vector\[i\] ^ 1}} |
| ColAndCol.java | {{vector1\[i\] & vector2\[0\]}} | {{vector1\[i\] & vector2Value}} |
| ColOrCol.java | {{vector1\[i\] \| vector2\[0\]}} | {{vector1\[i\] \| vector2Value}} |
| ColumnCompareColumn.txt | {{vector1\[i\] < vector2\[i\] ? 1 : 0}} | {{(vector1\[i\] - vector2\[i\]) & Long.MIN_VALUE}} |
| IfExprColumnColumn.txt | {{outputVector\[i\] = (vector1\[i\] == 1 ? vector2\[i\] : vector3\[i\])}} | {{outputVector\[i\] = (~(vector1\[i\] - 1) & vector2\[i\]) \| ((vector1\[i\] - 1) & vector3\[i\])}} |
| IfExprColumnScalar.txt | {{outputVector\[i\] = (vector1\[i\] == 1 ? arg2Scalar : vector3\[i\])}} | {{outputVector\[i\] = (~(vector1\[i\] - 1) & argScalar2) \| ((vector1\[i\] - 1) & vector3\[i\])}} |

Most of them are quite simple. May I make and start corresponding issues?

Sincerely yours,
Teddy Choi

> Optimization for SIMD instructions in Hive
> ------------------------------------------
>
>                 Key: HIVE-10179
>                 URL: https://issues.apache.org/jira/browse/HIVE-10179
>             Project: Hive
>          Issue Type: Improvement
>            Reporter: Chengxiang Li
>            Assignee: Chengxiang Li
>              Labels: optimization
>
> [SIMD|http://en.wikipedia.org/wiki/SIMD] instuctions could be found in most of current CPUs, such as Intel's SSE2, SSE3, SSE4.x, AVX and AVX2, and it would help Hive to outperform if we can vectorize the mathematical manipulation part of Hive. This umbrella JIRA may contains but not limited to the subtasks like:
> # Code schema adaption, current JVM is quite strictly on the code schema which could be transformed into SIMD instructions during execution. 
> # New implementation of mathematical manipulation part of Hive which designed to be optimized for SIMD instructions.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)