Figure 9-25 shows the response-time graph based on logical IO processing during our reporting interval. The response-time graphs for this example were created as described previously in this chapter, using OraPub's response-time graph template. The logical IO workload metric was chosen, since it typically has a high correlation to CPU consumption.

Because there is virtually no queue time related to servicing a logical IO, to reduce CPU consumption, we will need to reduce pure logical IO service time (typically, a buy strategy), the number of logical IO requests a SQL statement makes (tune strategy), or the rate of SQL statement executions (balance strategy). During the second analysis cycle, we will focus on this tuning approach.

Figure 9-26. This response-time graph focuses on the IO subsystem, so we chose physical IO as our workload metric. IO subsystems nearly always exhibit some queue time, and this situation is no different. Physical IO requests do include a significant amount of queue time, so we have multiple ways to reduce the physical IO-related response time. However, on this system, physical IOs are satisfied so quickly that the best course of action is to simply eliminate them by increasing the buffer cache.