Experimentation

After you read this document, you can better understand Capacity Advisor by experimenting, considering different configurations and workload placement, and by trying them out in what-if scenarios. A scenario identifies the workload demand profile that creates your experimental simulations.

When you do workload analysis on systems, you view graphs and reports that represent CPU or memory utilization by time. For example, Figure 2-2 shows a graph of CPU utilization for a single system over a one-month period.

Figure 2-2 CPU Utilization for Managed System puny03v8

Similarly, Figure 2-3 shows CPU utilization for a second system over the same period.

Figure 2-3 CPU Utilization for Managed System puny03v7

Comparing these two graphs shows that workload peaks on the two systems do not occur simultaneously, nor do they require the same percentage of the allocated CPU cores for processing. This suggests an opportunity to consider whether you can consolidate both systems together to satisfy the needs of the workloads, while reducing the number of CPU cores (originally each system is allocated 2 cores, for a total of 4 cores available to do work).

Figure 2-4 shows the result of using a Capacity Advisor “what-if” scenario to combine the workloads onto one system.

Figure 2-4 Combined What-If CPU Utilization for puny03v8 and puny03v7

From the graph, it is evident that the peak of the combined workloads is under 2 CPU cores. Even with utilization limits in place, this system is unlikely to need 4 CPU cores to meet this workload demand.