[
https://issues.apache.org/jira/browse/HBASE-16438?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15611207#comment-15611207
]
Anoop Sam John commented on HBASE-16438:
----------------------------------------
In CellChunkMap we write the index also to off heap chunks. So its ok to write
the seqId also along with chunkId+offset+length. So totally 20 bytes per cell
but all can be in offheap area. So we are in better place.
The other way of keeping the ref means we will need a ref array + the chunk. 2
data structures keep the index data. Per cell one entry to ref array (that is
on heap) and remaining 16 bytes can go to offheap chunk.
Only worry of using chunk from pool for the index is the wastage it can make.
May be we should share one chunk (One BB of 2 MB size) btw more than one
segment's index.
Elsewhere [[email protected]] was mentioning abt upping the chunk size from
2 MB too. Then the concern is more. With 2 MB we can keep 131072 cells index
data in it.
> Create a cell type so that chunk id is embedded in it
> -----------------------------------------------------
>
> Key: HBASE-16438
> URL: https://issues.apache.org/jira/browse/HBASE-16438
> Project: HBase
> Issue Type: Sub-task
> Affects Versions: 2.0.0
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
>
> For CellChunkMap we may need a cell such that the chunk out of which it was
> created, the id of the chunk be embedded in it so that when doing flattening
> we can use the chunk id as a meta data. More details will follow once the
> initial tasks are completed.
> Why we need to embed the chunkid in the Cell is described by [~anastas] in
> this remark over in parent issue
> https://issues.apache.org/jira/browse/HBASE-14921?focusedCommentId=15244119&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-15244119
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
