[ 
https://issues.apache.org/jira/browse/CASSANDRA-21178?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Stefan Miklosovic updated CASSANDRA-21178:
------------------------------------------
    Description: 
Rationale:

1) A user does not know when a training occurred so they can not evaluate if 
they shoul train again.
2) A user could implement "poor's man auto training" just by croning it. 
3) At the same time, we do not want to "over train", that is, we do not want to 
allow a user to train, for example, 100 times per minute as that does not make 
any sense. If we introduce created_at column with a timestamp, then we could 
say that not enough time elapsed since the last training and such operation 
would be failed.


  was:
Rationale:

1) A user does not know when a training occurred so they can not evaluate if 
they should not train again.
2) A user could implement "poor's man auto training" just by croning it. 
3) At the same time, we do not want to "over train", that is, we do not want to 
allow a user to train, for example, 100 times per minute as that does not make 
any sense. If we introduce created_at column with a timestamp, then we could 
say that not enough time elapsed since the last training and such operation 
would be failed.



> Introduce created_at column to system_distributed.compression_dictionaries
> --------------------------------------------------------------------------
>
>                 Key: CASSANDRA-21178
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21178
>             Project: Apache Cassandra
>          Issue Type: Task
>          Components: Feature/Compression
>            Reporter: Stefan Miklosovic
>            Assignee: Stefan Miklosovic
>            Priority: Normal
>             Fix For: 5.x
>
>
> Rationale:
> 1) A user does not know when a training occurred so they can not evaluate if 
> they shoul train again.
> 2) A user could implement "poor's man auto training" just by croning it. 
> 3) At the same time, we do not want to "over train", that is, we do not want 
> to allow a user to train, for example, 100 times per minute as that does not 
> make any sense. If we introduce created_at column with a timestamp, then we 
> could say that not enough time elapsed since the last training and such 
> operation would be failed.



--
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