[
https://issues.apache.org/jira/browse/HBASE-18010?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16053100#comment-16053100
]
Anastasia Braginsky commented on HBASE-18010:
---------------------------------------------
Thank you everyone for the comments so far!
bq. When big cells, what happens? Where/how will big cells be addressed?
The "big" cells are those bigger than the size of the chunk and thus not
allocated via MSLAB. In CellChunkMap everything must be on a chunk. One
possibility is to enlarge the chunk size. Currently there is an assert on cells
not allocated via MSLAB. This issue is going to be addressed just after I
finish with CellChunkMap integration. I have opened HBASE-18232 for that. This
JIRA is already big enough to add anything else to it.
bq. Did this issue get filed? Link it here if it did?
I didn't fix it here, because I was suggested to open a separate issue for
that. However, the fix is really small, [~stack], are you in favor of adding
the fix in this JIRA?
> Connect CellChunkMap to be used for flattening in CompactingMemStore
> --------------------------------------------------------------------
>
> Key: HBASE-18010
> URL: https://issues.apache.org/jira/browse/HBASE-18010
> Project: HBase
> Issue Type: Sub-task
> Reporter: Anastasia Braginsky
> Assignee: Anastasia Braginsky
> Attachments: HBASE-18010-V04.patch
>
>
> The CellChunkMap helps to create a new type of ImmutableSegment, where the
> index (CellSet's delegatee) is going to be CellChunkMap. No big cells or
> upserted cells are going to be supported here.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
