Before describing the problem I am facing, I would like to tell that I am a newbie in Teradata and started to learn recent days.
Problem statement : Since 1 month our Teradata server performance becoming very poor and we are experiencing slowness for the regular query results. As part of performance fine tuning, we have identified the most skewed tables without compression and did the compression (2TB to 1.5TB) and started modifying the PI as well.
Though we are trying to narrow down the issue we are unable to conclude whats going wrong inside the database. We have been working in this since 1 week and not able to conclude the factors which were involved to tweak the performance. What we have noticed is that whenever any one of the user query CPU skew reaches 90% and start continuing then the subsequent queries started to go to a long running state and not getting completed even after 6 hours !!!
Pls.note: The query which used to get executed in less than a mniute takes more than 20 minutes nowadays and we are not able to identify the root cause.
Could any one of you suggest or guide me to trouble shoot this issue in a proper channel.
Thanks and Regards,
Something to take with your tasm if this systemwide performance issue ? Can you check which of your workloads are taking how much cpu and then can start from there ? I am not an expert in tasm, but we had a similar issue and it turned out to be tasm.