[
https://issues.apache.org/jira/browse/CASSANDRA-15452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17823380#comment-17823380
]
Jon Haddad commented on CASSANDRA-15452:
----------------------------------------
I took a look at the new BTI format and we're doing the same thing here.
{noformat}
TIME COMM PID T BYTES OFF_KB LAT(ms) FILENAME
23:47:09 CompactionExec 44651 R 3982 0 0.01 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3982 0 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3980 3 0.01 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3980 3 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3986 7 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3986 7 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3986 11 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3986 11 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3982 15 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3982 15 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3982 19 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3982 19 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3989 23 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3989 23 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3996 27 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3996 27 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3974 31 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3974 31 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3987 35 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3987 35 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3984 38 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3984 38 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3973 42 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3973 42 0.00 da-30-bti-Data.db
23:47:09 CompactionExec 44651 R 3991 46 0.00
da-30-bti-Data.db{noformat}
We're also unnecessary
> Improve disk access patterns during compaction and streaming
> ------------------------------------------------------------
>
> Key: CASSANDRA-15452
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15452
> Project: Cassandra
> Issue Type: Improvement
> Components: Legacy/Local Write-Read Paths, Local/Compaction
> Reporter: Jon Haddad
> Priority: Normal
> Attachments: results.txt, sequential.fio
>
>
> On read heavy workloads Cassandra performs much better when using a low read
> ahead setting. In my tests I've seen an 5x improvement in throughput and
> more than a 50% reduction in latency. However, I've also observed that it
> can have a negative impact on compaction and streaming throughput. It
> especially negatively impacts cloud environments where small reads incur high
> costs in IOPS due to tiny requests.
> # We should investigate using POSIX_FADV_DONTNEED on files we're compacting
> to see if we can improve performance and reduce page faults.
> # This should be combined with an internal read ahead style buffer that
> Cassandra manages, similar to a BufferedInputStream but with our own
> machinery. This buffer should read fairly large blocks of data off disk at
> at time. EBS, for example, allows 1 IOP to be up to 256KB. A considerable
> amount of time is spent in blocking I/O during compaction and streaming.
> Reducing the frequency we read from disk should speed up all sequential I/O
> operations.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
