[
https://issues.apache.org/jira/browse/CASSANDRA-1531?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Stu Hood updated CASSANDRA-1531:
--------------------------------
Attachment: 0001-Optionally-include-indexes-when-creating-CFs-in-stre.patch
0002-Add-cardinality-option-and-use-it-to-write-determini.patch
0003-Add-IndexedRangeSlice-queries-each-distinct-value-in.patch
0001 Adds an option to create an index
0002 Makes column values deterministic, and unique based on a given cardinality
0003 Adds an 'indexedrangeslice' operation that divides the unique column
values between threads
The implementation of 0003 doesn't really use the range assigned to the thread,
except to determine how many index values to query for. All of the threads
start in the same base range, which is the worst case for a distributed
environment (on the other hand, that's probably the query people will be
running most in production).
I'll attach some initial benchmarks to CASSANDRA-1472 in a minute.
> Add secondary index ops to stress.py
> ------------------------------------
>
> Key: CASSANDRA-1531
> URL: https://issues.apache.org/jira/browse/CASSANDRA-1531
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Stu Hood
> Fix For: 0.7.0
>
> Attachments:
> 0001-Optionally-include-indexes-when-creating-CFs-in-stre.patch,
> 0002-Add-cardinality-option-and-use-it-to-write-determini.patch,
> 0003-Add-IndexedRangeSlice-queries-each-distinct-value-in.patch
>
>
> Secondary indexes need some basic benchmarks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
