hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sean Busbey (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17875) Document why objects over 10MB are not well-suited for hbase
Date Wed, 26 Apr 2017 21:32:04 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17875?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15985602#comment-15985602

Sean Busbey commented on HBASE-17875:

+Why are the cells above 10MB not recommended for HBase?::
+  The large cells cannot fit the buffer well in HBase. Large cells bypass the MemStoreLAB
per write operation and cannot be cached in the L2 block cache per read operation, HBase has
to allocate the on-heap memory once a time. The GC can be very bad.

Suggested rewrite:

"Large cells don't fit well into HBase's approach to buffering data. First, the large cells
bypass the MemStoreLAB when they are written. Then, they cannot be cached in the L2 block
cache during read operations. Instead, HBase has to allocate on-heap memory for them each
time. This can have a significant impact on the garbage collector within the RegionServer

> Document why objects over 10MB are not well-suited for hbase
> ------------------------------------------------------------
>                 Key: HBASE-17875
>                 URL: https://issues.apache.org/jira/browse/HBASE-17875
>             Project: HBase
>          Issue Type: Bug
>          Components: documentation, mob
>            Reporter: stack
>            Assignee: Jingcheng Du
>             Fix For: 2.0.0
>         Attachments: HBASE-17875.patch
> A new-user who has patently done their homework is unable to find in doc why objects
above 10MB are not recommended for hbase. Lets add explanation that has a link from MOB on
what happens when objects this size are requested form HBase.

This message was sent by Atlassian JIRA

View raw message