[ 
https://issues.apache.org/jira/browse/CASSANDRA-17021?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18030182#comment-18030182
 ] 

Jon Haddad edited comment on CASSANDRA-17021 at 10/15/25 9:49 PM:
------------------------------------------------------------------

I noticed that if you try to train a dictionary for a table but aren't using 
zstd dict compression, it still tries to start, then errors out like this:
{noformat}
bin/nodetool  traincompressiondictionary cassandra_easy_stress keyvalue
Starting compression dictionary training for cassandra_easy_stress.keyvalue...
Training from existing SSTables (flushing first if needed)
Failed to trigger training: null{noformat}
Can you add a check before it starts that makes sure the user is using a 
compression strategy that supports training?


was (Author: rustyrazorblade):
I noticed that if you try to train a dictionary for a table but aren't using 
zstd dict compression, it still tries to start, then errors out like this:


{noformat}
bin/nodetool  traincompressiondictionary cassandra_easy_stress keyvalue
Starting compression dictionary training for cassandra_easy_stress.keyvalue...
Training from existing SSTables (flushing first if needed)
Failed to trigger training: null{noformat}

Can you add a check that makes sure the user is using a compression strategy 
that supports training?

> Enhance Zstd support in Cassandra with dictionaries
> ---------------------------------------------------
>
>                 Key: CASSANDRA-17021
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-17021
>             Project: Apache Cassandra
>          Issue Type: Improvement
>          Components: Feature/Compression
>            Reporter: Dinesh Joshi
>            Assignee: Yifan Cai
>            Priority: Normal
>          Time Spent: 17h 20m
>  Remaining Estimate: 0h
>
> Currently Cassandra supports zstd compression. However, Zstd also supports 
> dictionaries to enhance not only the compression ratio but also the speed. 
> Dictionaries can show 3-4x savings. We should add support to train 
> dictionaries, ideally per SSTable this will yield the maximum gains.



--
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