[
https://issues.apache.org/jira/browse/HBASE-15554?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
ramkrishna.s.vasudevan updated HBASE-15554:
-------------------------------------------
Attachment: HBASE-15554_12.patch
Updated patch. This patch has the following updates
-> Now consumes ts and type part also
-> As per Anoop's suggestion - made one change to the handling of ROW_COL.
Previously I had assumed rowOFfset to be non zero and thus considering for pure
KV based Cells. But that is now changed so the extractByte() in
RowcolBloomExtractor has been slightly modified to handle all cases of cell.
-> Will raise another JIRA to remove the ts and type because that needs some
testing to confirm. It should work but still.
-> After changing the read path to work with Cell we can change the hash() algo
such that there is no offset needed and length can be calculated inside the
HashByteExtractor as everthing will be cell based. That can be done in another
JIRA.
> StoreFile$Writer.appendGeneralBloomFilter generates extra KV
> ------------------------------------------------------------
>
> Key: HBASE-15554
> URL: https://issues.apache.org/jira/browse/HBASE-15554
> Project: HBase
> Issue Type: Sub-task
> Components: Performance
> Reporter: Vladimir Rodionov
> Assignee: ramkrishna.s.vasudevan
> Fix For: 2.0.0
>
> Attachments: HBASE-15554.patch, HBASE-15554_10.patch,
> HBASE-15554_11.patch, HBASE-15554_12.patch, HBASE-15554_3.patch,
> HBASE-15554_4.patch, HBASE-15554_6.patch, HBASE-15554_7.patch,
> HBASE-15554_9.patch
>
>
> Accounts for 10% memory allocation in compaction thread when BloomFilterType
> is ROWCOL.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
