Workload Right Sizer
Workload Right Sizer analyzes CPU and memory in your virtualized environment and recommends tuning adjustments that you can execute via a script or change control.
Running Workload Right Sizer
Start by selecting the entity type to analyze: Application, ESX Cluster, ESX Host, or ESX VM, then select a named entity from the drop down list.
You can choose to use advanced options with Workload Right Sizer by checking the Use Advanced Options box. For example, you can change the CPU and memory over-subscription ratios, which are set to 2:1 by default.
You can select whether a higher tier VM gets precedence for resources if there is a conflict.
You can also set acceptable usage thresholds for vCPU and VM memory usage. This identifies candidates for right sizing.
You can also over-ride the thresholds for a specific tier. Say you wish to right size the Bronze tier differently than other tiers and select candidates based on different thresholds. You can do this by selecting “Override Thresholds by Tier”, selecting the tier and setting the usage thresholds to different values.
Set the date range for the analytic. When running Workload Right Sizer for the first time, choose a range where the entities have been in the same state. After the analytic is run for the first time, it’s advisable to run it monthly.
You can save analytics in the same way you can save a report and schedule them to run periodically. Save the analytic as a template and schedule it as a monthly job.
You can run the analytic in the background and view its output later on the Outputs page.
Understanding Workload Right Sizer Results
The top section shows you how many VMs were analyzed and a summary of the recommendations in terms of VM to grow, shrink, or a combination of both. Links to download the Right Sizer script or to submit a change request via the ServiceNow integration are also provided here.
The Virtual Machines tab shows a list of the analyzed VMs and recommendations for vCPU and memory. You can expand the results for each VM by clicking the down arrow next to its name. The results can be sorted by name, priority, or open change request.
Three additional tabs share analysis and recommendations grouped by hosts, clusters, and applications.
You can download a script to make the recommended changes.
You can also initiate and track ServiceNow change requests (requires the ServiceNow integration).
