[jira] [Updated] (CASSANDRA-15127) Add compression performance metrics

2019-06-04 Thread Jeremy Hanna (JIRA)


 [ 
https://issues.apache.org/jira/browse/CASSANDRA-15127?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Jeremy Hanna updated CASSANDRA-15127:
-
Component/s: Feature/Compression

> Add compression performance metrics
> ---
>
> Key: CASSANDRA-15127
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15127
> Project: Cassandra
>  Issue Type: Task
>  Components: Feature/Compression, Observability/Metrics
>Reporter: Michael
>Priority: Normal
>
> When doing some bulk load into my cluster I notice almost 100% CPU usage.
> As I am using DeflateCompressor, I assume that the data 
> compression/decompression contributes a lot to overall CPU load. 
> Unfortunately cassandra doesn't seem to have any metrics explaining how much 
> CPU time has been required for that.
> So I guess it would be useful to introduce cumulative times for compression 
> and decompression, breaking down by each supported compression algorithm.
> Then by comparing how much does each specific value increase per minute, with 
> number of processed requests and their cumulative times, it would be easy to 
> conclude how costly is the compression.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org



[jira] [Updated] (CASSANDRA-15127) Add compression performance metrics

2019-05-15 Thread Michael (JIRA)


 [ 
https://issues.apache.org/jira/browse/CASSANDRA-15127?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Michael updated CASSANDRA-15127:

Component/s: Observability/Metrics

> Add compression performance metrics
> ---
>
> Key: CASSANDRA-15127
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15127
> Project: Cassandra
>  Issue Type: Task
>  Components: Observability/Metrics
>Reporter: Michael
>Priority: Normal
>
> When doing some bulk load into my cluster I notice almost 100% CPU usage.
> As I am using DeflateCompressor, I assume that the data 
> compression/decompression contributes a lot to overall CPU load. 
> Unfortunately cassandra doesn't seem to have any metrics explaining how much 
> CPU time has been required for that.
> So I guess it would be useful to introduce cumulative times for compression 
> and decompression, breaking down by each supported compression algorithm.
> Then by comparing how much does each specific value increase per minute, with 
> number of processed requests and their cumulative times, it would be easy to 
> conclude how costly is the compression.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org