hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lokesh Jain (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-12521) Ozone: SCM should read all Container info into memory when booting up
Date Mon, 09 Oct 2017 10:12:00 GMT

    [ https://issues.apache.org/jira/browse/HDFS-12521?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16196785#comment-16196785

Lokesh Jain commented on HDFS-12521:

There are two issues which need to be solved through this jira :-
# ContainerStateManager should preload all the containers when booting up after restart.
# MetadataStoreBuilder should use caching provided by LevelDb and RocksDb for faster db operations.
LevelDb and RocksDb are internally configured to use a default cache of size 8 MB if a cache
object is not provided. We need to pass a cache object of the required capacity as an option.

> Ozone: SCM should read all Container info into memory when booting up
> ---------------------------------------------------------------------
>                 Key: HDFS-12521
>                 URL: https://issues.apache.org/jira/browse/HDFS-12521
>             Project: Hadoop HDFS
>          Issue Type: Sub-task
>          Components: ozone
>            Reporter: Anu Engineer
>            Assignee: Lokesh Jain
>              Labels: performance
> When SCM boots up it should read all containers into memory. This is a performance optimization
that allows delays on SCM side. This JIRA tracks that issue.

This message was sent by Atlassian JIRA

To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org

View raw message