cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kelvin Kakugawa (JIRA)" <>
Subject [jira] Commented: (CASSANDRA-1072) Increment counters
Date Tue, 30 Nov 2010 18:06:20 GMT


Kelvin Kakugawa commented on CASSANDRA-1072:

We've also tried to test 1546, internally, but at scale it's counter-specific read/write code
paths add too much contention to the thread stages.

Consequently, we also invested a reasonable effort into optimizing 1546: removed a deadlock,
found an unnecessary lock around SST compaction.  However, ultimately, we re-did the logic
in 1072 to avoid the lock strategy of 1546.  1546's additional, counter-specific read/write
paths require more reasoning about how it interacts w/ the underlying system.  Whereas, 1072's
extension is more limited in scope, because it only extends the basic logic of Column classes.

> Increment counters
> ------------------
>                 Key: CASSANDRA-1072
>                 URL:
>             Project: Cassandra
>          Issue Type: Sub-task
>          Components: Core
>            Reporter: Johan Oskarsson
>            Assignee: Kelvin Kakugawa
>         Attachments: CASSANDRA-1072.112210.patch, CASSANDRA-1072.patch,,
> Break out the increment counters out of CASSANDRA-580. Classes are shared between the
two features but without the plain version vector code the changeset becomes smaller and more

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

View raw message