[
https://issues.apache.org/jira/browse/CASSANDRA-21632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Brandon Williams updated CASSANDRA-21632:
-----------------------------------------
Fix Version/s: 5.0.10
6.0-alpha3
7.0
(was: 5.0.x)
(was: 7.x)
(was: 6.x)
Since Version: 5.0-alpha1
Source Control Link:
https://github.com/apache/cassandra/commit/e8b7eac7ed1e62a8d3cccb014c6c4b3f9f0a33c4
Resolution: Fixed
Status: Resolved (was: Ready to Commit)
Committed, thanks!
> BTI partition index preload is unbounded
> ----------------------------------------
>
> Key: CASSANDRA-21632
> URL: https://issues.apache.org/jira/browse/CASSANDRA-21632
> Project: Apache Cassandra
> Issue Type: Bug
> Components: Local/SSTable, Local/Startup and Shutdown
> Reporter: Alexandre Viau
> Assignee: Alexandre Viau
> Priority: Normal
> Labels: patch-available
> Fix For: 5.0.10, 6.0-alpha3, 7.0
>
> Attachments: ci_summary_apache_5.0_199.html,
> ci_summary_apache_6.0_139.html, ci_summary_apache_trunk_696.html,
> results_details_apache_5.0_199.tar.xz, results_details_apache_6.0_139.tar.xz,
> results_details_apache_trunk_696.tar.xz
>
> Time Spent: 2.5h
> Remaining Estimate: 0h
>
> When a table has bloom_filter_fp_chance = 1, PartitionIndex.load warms the
> entire Partitions.db by seeking to every 4096-byte offset and reading a
> single byte. The warm is unbounded. Nothing relates the amount read to the
> amount of page cache the node can actually retain
>
> On our cluster (5.4 TiB and 32.5B partitions per node, bloom_filter_fp_chance
> = 1, Partitions.db totalling 309.47 GiB across 798 files, or 81,125,849 pages
> to touch), roughly 70 GiB of page cache is available to the process. The warm
> reads 309 GiB to retain at most a fraction of it, and startup takes roughly
> 90 minutes, essentially all of it in SSTableReader.openAll.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
