By: Ryan Cartwright & Mark Monce, VMware Senior Systems Engineers
On May 11th, we will be hosting a Getting More Out of VMware webinar about Optimizing your Existing Capacity using vRealize Operations (vR Ops). It will be structured around eight steps that I’d like to preview for you now.
Effective and consistent vSphere capacity management is essential to getting the most value from your vSphere infrastructure. VMware vRealize Operations 6 provides the insight and information you and your managers need to get the maximum use and efficiency from your virtualization resources.
The fact is that different consumption models exist. It’s important to understand the various models and to know that there is no right or wrong approach. However, this conversation becomes an important step in the process to determine your organization’s approach to make your capacity planning and management decisions. Which metric should I be using to monitor CPU (Demand vs Allocation), Memory (Demand vs Allocation vs Consumed), and Disk Space (Demand vs Allocation)? Different customers approach capacity management in different ways for their different workloads.
Most importantly, these analyses must reflect the business priorities and management philosophy of your organization. Within the Administration, you should view, modify, and create custom policies to reflect your organization’s consumption model. Remember, all workloads are not created equal and should not be treated equal. Different policies can be set up for Business Critical Apps, Production, vs Test & Dev. You can modify settings that affect Capacity Remaining and Time Remaining, and Reclaimable Capacity badges. These policies can be applied to specific custom groups to affect different groups of objects. For additional information on custom groups, see our Getting More Out of VMware Webinar on Groups and Policies from July 22, 2015.
Capacity Remaining Badge
When planning your resources, you must understand the quantity or capacity of resources available for memory, CPU, and disk space for a specific object whether that is a capacity container or workload. The capacity remaining badge represents the remaining capability of your environment to accommodate new virtual machines. By default, we take into consideration HA settings as well as additional 10% buffer which can be modified in the policy.
Time Remaining Badge
Time remaining is a measure of the amount of time left before your objects run out of capacity. You must be able to translate that into the amount of time you have to respond. The Time Remaining badge indicates how much time is remaining before the resources of the object exhaust. By default, we have a provisioning buffer of 30 days. If any of the compute resources has less capacity than the provisioned buffer, the Time Remaining score is 0.
Reclaimable Capacity Badge
Now that you have an accurate picture of the current situation and future trends, you can then turn this information into action. For example, can you reclaim existing infrastructure before acquiring and deploying more? Reclaimable capacity is the amount of provisioned capacity that be reclaimed without causing stress or performance degradation. Reclaimable Capacity is calculated for each resource type like CPU, memory, and disk, for each object in the environment. It identifies the amount of resources that can be reclaimed and provisioned to other objects in your environment.
You can use the vRealize Operations Manager projects feature to plan for capacity upgrades or optimization in your virtual environment. You create a project to include upcoming changes that affect your capacity and is useful for capacity planning. A project has one or more scenarios where capacity or demand changes on a date in the future. You can visualize one or more projects in the visualization pane to forecast your capacity and demand. When a project is expected to happen, you can commit it, which reserves the capacity needed and updates the capacity and/or time remaining badges.
Finally, understanding how your current usage of vSphere itself can impact capacity—and possibly offer resource savings—is important. In large environments, proper understanding of the basics of Distributed Resource Scheduler will help you analyze and optimize resource utilization through proper placement. For more information on DRS, join the upcoming Getting More Out of VMware Webinar on Distributed Resource Management on June 8, 2016.
In vRealize Operations Manager 6.2, a new Workload Utilization Dashboard to visualize your object utilization across your hosts and clusters. This can be customized to show any type of object within your environment. There is a Balance action that can be used to distribute your workloads across clusters to help distribute your workloads based being bound by CPU or Memory that can be leveraged manually or automated.
See Getting More Out of VMware Webinar on Automated Actions, Load Balancing, and Workload Placement from September 23, 2015 for additional information.
Hopefully this preview has piqued your interest in joining us on May 11th, when we’ll go into more depth about Optimizing your Existing Capacity for vRealize Operations. Click here to register.
Visit www.vmware.com/go/getmore to view the entire Getting More Out of VMware webinar series.