AngersZhuuuu commented on pull request #34326:
URL: https://github.com/apache/spark/pull/34326#issuecomment-1006317362


   > It is great to see that more metrics are being added to History server. 
Updating the listing.ldb is one of the major factors impacting the History 
server performance, there are several new metrics tracked for listing.ldb. But 
there is also another factor, which is generating each individual application 
levelDB when there is a user request. There are some metrics tracking the 
LevelDB generating time and loading time in this PR. Can you help add some more 
metrics, such as how many applications are being parsed to generated their own 
LevelDBs, and how many LevelDBs are currently stored in Disk, and what is the 
total size of them. Thanks.
   
   Update a change. And I want confirm with you that:
   1. If we don't set `spark.history.store.path`, it will use memory to store 
app info and it's not belong to LevelDB metrics, right?
   2. Only when support hybrid store disk backend,  then we have in-memory 
LevelDBs, right?


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to