You are viewing a plain text version of this content. The canonical link for it is here.

Posted to jira@arrow.apache.org by "zhaoyaqi (Jira)" <ji...@apache.org> on 2022/09/29 02:33:00 UTC

[jira] [Assigned] (ARROW-17884) Add Intel®-IAA/QPL-based Parquet RLE Decode

     [ https://issues.apache.org/jira/browse/ARROW-17884?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

zhaoyaqi reassigned ARROW-17884:
--------------------------------

    Assignee: zhaoyaqi

> Add Intel®-IAA/QPL-based Parquet RLE Decode
> -------------------------------------------
>
>                 Key: ARROW-17884
>                 URL: https://issues.apache.org/jira/browse/ARROW-17884
>             Project: Apache Arrow
>          Issue Type: Improvement
>          Components: C++
>            Reporter: zhaoyaqi
>            Assignee: zhaoyaqi
>            Priority: Minor
>              Labels: performance, pull-request-available
>          Time Spent: 20m
>  Remaining Estimate: 0h
>
> Intel® In-Memory Analytics Accelerator (Intel® IAA) is a hardware accelerator available in the upcoming generation of Intel® Xeon® Scalable processors ("Sapphire Rapids"). Its goal is to speed up common operations in analytics like data (de)compression and filtering. It support decoding of Parquet RLE format. We add new codec which utilizes the Intel® IAA offloading technology to provide a high-performance RLE decode implementation. The codec uses the [Intel® Query Processing Library (QPL)|https://github.com/intel/qpl] which abstracts access to the hardware accelerator. The new solution provides in general higher performance against current solution, and also consume less CPU.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)