Most of the time, we use wait stats to identify SQL Server’s bottleneck and focus on what to tune.
However, let’s say you’ve got a really well-tuned SQL Server where:
Queries finish in milliseconds (think 5-20 milliseconds, super quick) Everything your queries need is cached in memory There are no blocking issues The workload doesn’t do a lot of writes, and the few writes that it does, hit the transaction log extremely quickly because you’ve got great storage
In that situation, you may not see any significant wait stats, but as your workload grows, your CPU usage will