[
https://issues.apache.org/jira/browse/CASSANDRA-15393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16964267#comment-16964267
]
Blake Eggleston edited comment on CASSANDRA-15393 at 10/31/19 6:07 PM:
-----------------------------------------------------------------------
bq. Would the local-read benefits be better served by removing ByteBuffer
entirely?
Could you clarify that? Do you mean remove ByteBuffer from the cell interface
entirely and using byte arrays for everything?
The linked spreadsheet has 3 tabs, the "Tiny" tab you're probably looking at is
just measuring the {{(k int primary key, v int)}} data model. The medium and
wide tabs are workloads represent more common workloads.
was (Author: bdeggleston):
> Would the local-read benefits be better served by removing ByteBuffer
> entirely?
Could you clarify that? Do you mean remove ByteBuffer from the cell interface
entirely and using byte arrays for everything?
The linked spreadsheet has 3 tabs, the "Tiny" tab you're probably looking at is
just measuring the {{(k int primary key, v int)}} data model. The medium and
wide tabs are workloads represent more common workloads.
> Add byte array backed cells
> ---------------------------
>
> Key: CASSANDRA-15393
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15393
> Project: Cassandra
> Issue Type: Sub-task
> Components: Local/Compaction
> Reporter: Blake Eggleston
> Assignee: Blake Eggleston
> Priority: Normal
> Fix For: 4.0
>
>
> We currently materialize all values as on heap byte buffers. Byte buffers
> have a fairly high overhead given how frequently they’re used, and on the
> compaction and local read path we don’t do anything that needs them. Use of
> byte buffer methods only happens on the coordinator. Using cells that are
> backed by byte arrays instead in these situations reduces compaction and read
> garbage up to 22% in many cases.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
