[SERVER-73620] Report in serverStatus the amount of data written to disk by query system Created: 04/Feb/23  Updated: 28/Feb/23

Status: Backlog
Project: Core Server
Component/s: Diagnostics
Affects Version/s: None
Fix Version/s: None

Type: New Feature Priority: Major - P3
Reporter: Arun Banala Assignee: Backlog - Query Execution
Resolution: Unresolved Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Assigned Teams:
Query Execution
Participants:

 Description   

Currently if there is heavy disk usage by the server, there is no easy way to track if the IO is being caused by write operations or if a query is consuming the disk. Similar to SERVER-64660, we should also track the amount of the data consumed.



 Comments   
Comment by Ana Meza [ 28/Feb/23 ]

Sending it to backlog since we are not working on this for now

Comment by John Murphy [ 23/Feb/23 ]

Hi arun.banala@mongodb.com 

We have WiredTiger metrics that show how much is read/written to cache, along with the raw block device metrics that show how much is actually read/written to disk.

If my understanding is correct, the thing we don't have is how much disk IO is performed by the External Sorter when it spills to disk. I believe that if we had this data available to us, it would have been much easier to determine that the multi-planner was generating a high amount of sort operations that were spilling to disk and causing high disk utilisation.

Generated at Thu Feb 08 06:25:11 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.