You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@parquet.apache.org by "Wes McKinney (JIRA)" <ji...@apache.org> on 2018/11/11 22:19:00 UTC

[jira] [Commented] (PARQUET-517) [C++] Use arrow::MemoryPool for all heap allocations

    [ https://issues.apache.org/jira/browse/PARQUET-517?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16683042#comment-16683042 ] 

Wes McKinney commented on PARQUET-517:
--------------------------------------

I renamed the issue. Using {{arrow::MemoryPool}} for allocations will ensure alignment

> [C++] Use arrow::MemoryPool for all heap allocations
> ----------------------------------------------------
>
>                 Key: PARQUET-517
>                 URL: https://issues.apache.org/jira/browse/PARQUET-517
>             Project: Parquet
>          Issue Type: Improvement
>          Components: parquet-cpp
>            Reporter: Wes McKinney
>            Priority: Major
>             Fix For: cpp-1.6.0
>
>
> We are using {{std::vector}} in many places for memory allocation; if we want to use SSE on this memory we may run into some problems.
> Couple things we should do
> * Add an STL allocator for {{std::vector}} that ensure 16-byte aligned memory
> * Check user-provided memory for alignment before utilizing an SSE-accelerated routine (e.g. SSE hash functions for dictionary encoding) and decide whether to copy and use SSE or no-copy and use no-SSE code.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)