You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@geode.apache.org by "paul nikonowicz (Jira)" <ji...@apache.org> on 2021/02/01 21:34:00 UTC

[jira] [Comment Edited] (GEODE-4658) Expose how much time it takes to write to disk and what is the disk store size

    [ https://issues.apache.org/jira/browse/GEODE-4658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17276672#comment-17276672 ] 

paul nikonowicz edited comment on GEODE-4658 at 2/1/21, 9:33 PM:
-----------------------------------------------------------------

Should we aggregate metrics results for DiskStore? There can be many disk stores to report metrics on.

 

In order to get the disk stores, we need to know which disk stores to retrieve metrics for. For instance:



 
{code:java}
DiskStoreMXBean diskStoreMXBean = managementService.getLocalDiskStoreMBean(diskStore.getName());
 
{code}
 

Therefore, disk store metrics may need to be displayed as an aggregate sum/average or via some other display property that would make sense in the results table:
{code:java}
Category  |        Metric         | Value
--------- | --------------------- | -----
cluster   | totalHeapSize         | 702
cache     | totalRegionEntryCount | 0
          | totalRegionCount      | 1
          | totalMissCount        | 0
          | totalHitCount         | 0
diskstore | totalDiskUsage        | 0
          | diskReadsRate         | 0.0
          | diskWritesRate        | 0.0
          | flushTimeAvgLatency   | 0
          | totalBackupInProgress | 0
query     | activeCQCount         | 0
          | queryRequestRate      | 0.0
{code}


was (Author: pnikonowicz):
Should we aggregate metrics results for DiskStore? There can be many disk stores to report metrics on.

 

In order to get the disk stores, we need to know which disk stores to retrieve metrics for. For instance:
```

DiskStoreMXBean diskStoreMXBean = managementService.getLocalDiskStoreMBean(diskStore.getName());

```




Therefore, disk store metrics may need to be displayed as an aggregate sum/average or via some other display property that would make sense in the results table:
```

Category | Metric | Value
--------- | --------------------- | -----
cluster | totalHeapSize | 702
cache | totalRegionEntryCount | 0
 | totalRegionCount | 1
 | totalMissCount | 0
 | totalHitCount | 0
diskstore | totalDiskUsage | 0
 | diskReadsRate | 0.0
 | diskWritesRate | 0.0
 | flushTimeAvgLatency | 0
 | totalBackupInProgress | 0
query | activeCQCount | 0
 | queryRequestRate | 0.0

```

> Expose how much time it takes to write to disk and what is the disk store size
> ------------------------------------------------------------------------------
>
>                 Key: GEODE-4658
>                 URL: https://issues.apache.org/jira/browse/GEODE-4658
>             Project: Geode
>          Issue Type: Improvement
>          Components: docs, gfsh
>            Reporter: Fred Krone
>            Assignee: Nick Vallely
>            Priority: Major
>              Labels: starter++
>
> *Given* I have a persistent region with a default disk store
> *When* I write data on the region
> *Then* I can see  2 new metrics {{average write time to disk}} ns time and {{size of disk store}} in bytes
> h3. Documentation
> Add these new metrics to the docs.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)