[
https://issues.apache.org/jira/browse/CASSANDRA-4885?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13579203#comment-13579203
]
Jason Brown commented on CASSANDRA-4885:
----------------------------------------
Patch 4885-v1: Added a new field in Descriptor to declare if the sstable
version contains an extra byte per row (which declares if the row will have a
column-level bloom filter). Main functionality for writing or not writing out
bf lies in the RIE.serialize()/deserialize(), and ColumnIndex.Builder. Most
other changes are tooling support (cql3, cqlsh, and cli), and the attendent
changes in avro and thrift.
Some of the the bulk in the patch (oh, maybe 60% or so) is just the diff in the
auto-generated thrift classes. Wasn't sure if I should have excluded them or
not.
> Remove or rework per-row bloom filters
> --------------------------------------
>
> Key: CASSANDRA-4885
> URL: https://issues.apache.org/jira/browse/CASSANDRA-4885
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jonathan Ellis
> Assignee: Jason Brown
> Fix For: 2.0
>
> Attachments: 0001-CASSANRDA-4885-Remove-per-row-bloom-filter.patch,
> 0002-CASSANRDA-4885-update-test.patch, 4885-v1.patch
>
>
> Per-row bloom filters may be a misfeature.
> On small rows we don't create them.
> On large rows we essentially only do slice queries that can't take advantage
> of it.
> And on very large rows if we ever did deserialize it, the performance hit of
> doing so would outweigh the benefit of skipping the actual read.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
