Aws Aurora RDS Mysql instance unusual cpu spike pattern

yeah, I am investigating that as well but there some query which is from INFORMATION_SCHEMA which is taking high cpu sometimes , there is mine main concern because till now didn't notice that kind query taking that much cpu

SELECT LOGFILE_GROUP_NAME , FILE_NAME , TOTAL_EXTENTS , INITIAL_SIZE , ENGINE , EXTRA FROM INFORMATION_SCHEMA . FILES WHERE FILE_TYPE = ? AND FILE_NAME IS NOT NULL AND LOGFILE_GROUP_NAME IS NOT NULL AND LOGFILE_GROUP_NAME IN ( SELECT DISTINCTROW LOGFILE_GROUP_NAME FROM INFORMATION_SCHEMA . FILES WHERE FILE_TYPE = ? AND TABLESPACE_NAME IN ( SELECT DISTINCTROW TABLESPACE_NAME FROM INFORMATION_SCHEMA . PARTITIONS WHERE TABLE_SCHEMA IN (?) ) ) GROUP BY `LOGFILE_

Did you recently upgrade DB Engines by chance? Such as from MySQL 5.6 to MySQL 5.7? There are some difference in how the two choose query executions plans. So if you recently upgrade you may want to check that.

Hi! Did you resolve this issue? If yes, could you tell me how?

If not, here is my issue and a finding to help you troubleshoot yours. Currently, I am facing something like that. I was using an RDS Postgresql instance, and it was working well. But, after migrating to Aurora, we started to face some CPU spikes. We investigated all queries, indexes, application code ... everything. As an ultimate resource, we created another RDS instance to see if the behavior would be the same and, to our surprise, the spikes were gone. So, now, we are trying to figure out what configurations we should change/set to make it work on Aurora. It seems to be a widespread problem with RDS and Aurora. A quick search on Google brings a lot of questions throughout forums about it.