[ 
https://issues.apache.org/jira/browse/CASSANDRA-21134?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18057927#comment-18057927
 ] 

Sam Lightfoot edited comment on CASSANDRA-21134 at 2/11/26 5:27 PM:
--------------------------------------------------------------------

Preliminary test results for unthrottled cursor compaction with direct IO for 
compaction writes:
 * Memory: 12GB
 * Max Heap: 6GB
 * Hot dataset size: 1GB
 * Compaction reads: buffered

!image-2026-02-11-17-22-58-361.png|width=497,height=348!

!image-2026-02-11-17-25-58-329.png|width=291,height=182!

| Metric | Buffered (avg) | Direct IO (avg) | Improvement |
|---|---|---|---|
| **p99 read latency** | 16.2 ms | 1.9 ms | **8.5x** |
| **p99.9 read latency** | 100.6 ms | 7.9 ms | **12.7x** |
| **Mean read latency** | 1.29 ms | 0.61 ms | **2.1x** |
| **Stall_us/s** | 31,599 | 21,534 | **-32%** |
| **Active_File** | 471 MB | 488 MB | **+4%** |
| **Compaction throughput** | 246 MiB/s | 286 MiB/s | **+16%** |
| p50 read latency | 0.48 ms | 0.48 ms | — |

 


was (Author: JIRAUSER302824):
Preliminary test results for unthrottled cursor compaction with direct IO for 
compaction writes:
 * Memory: 12GB
 * Max Heap: 6GB
 * Hot dataset size: 1GB
 * Compaction reads: buffered

!image-2026-02-11-17-22-58-361.png|width=497,height=348!

> Direct IO support for compaction writes
> ---------------------------------------
>
>                 Key: CASSANDRA-21134
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21134
>             Project: Apache Cassandra
>          Issue Type: Improvement
>          Components: Local/Compaction
>            Reporter: Sam Lightfoot
>            Assignee: Sam Lightfoot
>            Priority: Normal
>             Fix For: 5.x
>
>         Attachments: image-2026-02-11-17-22-58-361.png, 
> image-2026-02-11-17-25-58-329.png
>
>
> Follow-up from the implementation for compaction reads (CASSANDRA-19987)
> Notable points
>  * Update the start-up check that impacts DIO writes 
> ({_}checkKernelBug1057843{_})
>  * RocksDB uses 1 MB flush buffer. This should be configurable and 
> performance tested (256KB vs 1MB)
>  * Introduce compaction_write_disk_access_mode
>  * Support for the compressed path would be most beneficial



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to