An Azure relational database service.
SQL Server instance CPU persistently high
We have a database pool that has been running well, however since ~13 May the pool has been consistently high CPU. There is a lot of noise of legit queries, so I am showing smaller windows. The last hour looks like this.
Zooming to a similar time period last Saturday
I have reviewed the pool to see which databases are using activity - the biggest consistent CPU is the hangfire databases.
The thing I don't understand with this is we haven't changed anything related to hangfire, the number of jobs is fairly constant, I did see some improvement in the stats when I deleted some successful jobs.
This is the query performance insight - it shows the hangfire query causing the CPU, which jumped up and remains high.
I'm not sure how to identify the cause, or whether there is an issue with the pool. Is there a way to identify what the instance might be doing that we can't see?