[
https://issues.apache.org/jira/browse/CASSANDRA-15452?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17943517#comment-17943517
]
Benedict Elliott Smith commented on CASSANDRA-15452:
----------------------------------------------------
While I'm looking at it, without wanting to totally reopen the patch, I wonder
if somebody could explain the reasoning behind using a ThreadLocal Map for
storing these blocks while they're in use? We seem to remove them as soon as we
close, and we can't use it in more than one ReadAheadBuffer, so what are we
gaining by using the ThreadLocal and Map? Why don't we just use a field in the
ReadAheadBuffer?
> Improve disk access patterns during compaction and range reads
> --------------------------------------------------------------
>
> Key: CASSANDRA-15452
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15452
> Project: Apache Cassandra
> Issue Type: Improvement
> Components: Legacy/Local Write-Read Paths, Local/Compaction
> Reporter: Jon Haddad
> Assignee: Jordan West
> Priority: Normal
> Fix For: 5.0.4, 5.1
>
> Attachments: ci_summary_jrwest_jwest-15452-5.0_153.html, everyfs.txt,
> image-2024-11-22-16-17-23-194.png, image-2025-01-07-16-04-23-909.png,
> image-2025-01-07-16-56-12-853.png, image-2025-01-07-16-57-29-134.png,
> iostat-5.0-head.output, iostat-5.0-patched.output, iostat-ebs-15452.png,
> iostat-ebs-head.png, iostat-instance-15452.png, iostat-instance-head.png,
> results.txt, results_details_jrwest_jwest-15452-5.0_153.tar.xz,
> screenshot-1.png, screenshot-2.png, screenshot-3.png, screenshot-4.png,
> screenshot-5.png, screenshot-6.png, sequential.fio, throughput-1.png,
> throughput.png
>
> Time Spent: 9h 40m
> Remaining Estimate: 0h
>
> 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.
> # We can reduce system calls by buffering writes as well, but I think it
> will have less of an impact than the reads
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
