[
https://issues.apache.org/jira/browse/HBASE-14920?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15271382#comment-15271382
]
Eshcar Hillel commented on HBASE-14920:
---------------------------------------
A core property of the new in-memory flush and compaction is retaining data in
memory for longer time. In the majority of cases flush to disk is invoked to
reduce memory usage. In less frequent use cases the user wish to clear the
memory.
If we only consider the second case and flush the entire content to disk with
every flush request, we significantly weaken the feature strength.
Alternatively, we can have this feature well documented, so that user
understand that if they wish to clear the memory they may need to invoke flush
several times.
Actually the same can happen for a default memstore.
Consider the case where while a flush to disk is in progress the user invokes a
flush to clear memory. However when the user flush is executed it finds the
snapshot segment still has entries -- it only logs a warning and ignores the
user flush request. So in this case as well a user request is not satisfied,
and the user should be able to handle this.
> Compacting Memstore
> -------------------
>
> Key: HBASE-14920
> URL: https://issues.apache.org/jira/browse/HBASE-14920
> Project: HBase
> Issue Type: Sub-task
> Reporter: Eshcar Hillel
> Assignee: Eshcar Hillel
> Attachments: HBASE-14920-V01.patch, HBASE-14920-V02.patch,
> HBASE-14920-V03.patch, HBASE-14920-V04.patch, HBASE-14920-V05.patch,
> HBASE-14920-V06.patch, HBASE-14920-V07.patch, move.to.junit4.patch
>
>
> Implementation of a new compacting memstore with non-optimized immutable
> segment representation
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
