You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Jim Witschey (JIRA)" <ji...@apache.org> on 2016/01/27 20:58:40 UTC

[jira] [Created] (CASSANDRA-11081) Benchmark for & write recommendations for which SSTable compressor to choose

Jim Witschey created CASSANDRA-11081:
----------------------------------------

             Summary: Benchmark for & write recommendations for which SSTable compressor to choose
                 Key: CASSANDRA-11081
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-11081
             Project: Cassandra
          Issue Type: Task
          Components: Documentation and Website
            Reporter: Jim Witschey
            Priority: Minor


This task was generated in breaking apart the initial test plan for CASSANDRA-10995. [~slebresne] said doing this task would be useful, but out of scope for that ticket, and perhaps not worth the effort right now.

It would be helpful to us and for users to know: what kinds of workloads should use which SSTable compressor? Currently, I believe all we can do for users is say "run your own benchmarks", but it would be good to have enough information to say, e.g., "workloads with (property x) benefit from using LZ4". Post-CASSANDRA-8099, there is less on-disk storage engine overhead that can be compressed, so choice of compressor is probably more workload-sensitive than it used to be, so the choice of using compression in the first place is less obvious.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)