Varun Saxena commented on YARN-4053:

This patch demonstrates the approach mentioned above and works for both 
integral and floating point values. But for floating point values, the 
restriction on part of the client is that it should send values in decimal 
format always otherwise when I add metric filters, matching will fail. I guess 
its a fair enough restriction to place.

In the patch, we can indicate that numerical values have to be stored per 
column/column prefix.
We can however extend this logic for all values and indicate if values to be 
stored are ASCII encoded as well, so that different kind of values can be 
stored differently in same column.

But there is no use case for this as of now, so haven't done so. 

I will remove the part about floating point numbers from patch, if we dont want 
it now.

> Change the way metric values are stored in HBase Storage
> --------------------------------------------------------
>                 Key: YARN-4053
>                 URL: https://issues.apache.org/jira/browse/YARN-4053
>             Project: Hadoop YARN
>          Issue Type: Sub-task
>          Components: timelineserver
>    Affects Versions: YARN-2928
>            Reporter: Varun Saxena
>            Assignee: Varun Saxena
>         Attachments: YARN-4053-YARN-2928.01.patch
> Currently HBase implementation uses GenericObjectMapper to convert and store 
> values in backend HBase storage. This converts everything into a string 
> representation(ASCII/UTF-8 encoded byte array).
> While this is fine in most cases, it does not quite serve our use case for 
> metrics. 
> So we need to decide how are we going to encode and decode metric values and 
> store them in HBase.

This message was sent by Atlassian JIRA

Reply via email to