hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Wei Zheng (JIRA)" <>
Subject [jira] [Commented] (HIVE-10233) Hive on tez: memory manager for grace hash join
Date Tue, 23 Jun 2015 18:13:01 GMT


Wei Zheng commented on HIVE-10233:

[~hagleitn] [~vikram.dixit]
1. Same question as [~mmokhtar] mentioned, why do we only allocate less than 1% memory to
the mapjoin?
2. What is the use for pctx.getConf(); in the beginning of MemoryDecider.resolve()?
3. For these three method calls, the param work is not used, so can be removed. Also can consider
removing the param work in evaluateWork(TezWork work, BaseWork w).
evaluateMapWork(work, (MapWork) w);
evaluateReduceWork(work, (ReduceWork) w);
evaluateMergeWork(work, (MergeJoinWork) w);
4. In evaluateOperators(BaseWork w, PhysicalContext pctx), pctx is not used
5. Indentation for evaluateOperators needs to be adjusted

> Hive on tez: memory manager for grace hash join
> -----------------------------------------------
>                 Key: HIVE-10233
>                 URL:
>             Project: Hive
>          Issue Type: Bug
>          Components: Tez
>    Affects Versions: llap, 2.0.0
>            Reporter: Vikram Dixit K
>            Assignee: Gunther Hagleitner
>         Attachments: HIVE-10233-WIP-2.patch, HIVE-10233-WIP-3.patch, HIVE-10233-WIP-4.patch,
HIVE-10233-WIP-5.patch, HIVE-10233-WIP-6.patch, HIVE-10233-WIP-7.patch, HIVE-10233-WIP-8.patch,
HIVE-10233.08.patch, HIVE-10233.09.patch, HIVE-10233.10.patch, HIVE-10233.11.patch
> We need a memory manager in llap/tez to manage the usage of memory across threads. 

This message was sent by Atlassian JIRA

View raw message