vROps – Workload Balance | Datacenter Resource Management | VMware vSphere Central

vROps – Workload Balance

image

This walkthrough is designed to provide a step-by-step demonstration of VMware vRealize Operations Manager’s Workload Placement - Rebalance feature. Use the arrow keys to navigate through the screens.

image

Workloads are often placed sub-optimally, resulting in inefficient usage of clusters and chronic resource contention. Resolving contention by balancing resources is a difficult, time-consuming, and non-strategic activity. To solve this, we need to visualize how the workloads are distributed and utilized across the environment.

image

Distributed Resource Scheduler (DRS) is a well-known and proven vSphere feature that moves VMs within a cluster to ensure VMs are always running on a host with adequate resources to support it. vRealize Operations Manager’s Rebalance feature moves VMs between clusters to ensure the clusters are balanced in the environment, which in the end helps DRS.

image

Let's start by logging in to the vRealize Operations Server. You can do this by pointing your web browser to https://

image

Now that we're logged in, we will want to navigate to the "Workload Balance" dashboard. To do this, we will start by expanding the side menu.

image

Then, we will continue by selecting the "Workload Balance" dashboard from the list of available dashboards.

image

We are now greeted by the "Workload Balance" dashboard. Here we can see multiple visual representations of how balanced the clusters are within the selected Datacenter. In this example, we have a single Datacenter by the name of "DemoDC". The first section of this dashboard is the "level" showing the overall balance of the available clusters. This dashboard also provides how many vMotions have occurred within a period of time, and a visualization of each cluster's workloads as well as their current DRS configuration. If we scroll down we can see additional information.

image

The "Workload Balance" dashboard also shows a capacity utilization visualization across many resources within the selected datacenter. By default, this chart represents what workload is the most constrained. We can also customize these charts by selecting the "Constrained By" dropdown box.

image

The workloads available include host, network, storage, and VM based metrics. Selecting any of those will modify the charts based on the metric selected.We also have the ability to interact with the individual resources simply by clicking on one of them.

image

After clicking on one of the resources, we can see some more detailed information. We see that the overall workload for this ESXi host is 88% and memory is the most constrained workload. To gain further information, you could click on "Details" where we would be taken to that ESXi host's Workload Analysis page. However, instead, we will scroll further down to see the last area of this dashboard.

image

The last area of the "Workload Balance" dashboard is the "Workload Trend" area. Here we see the CPU and Memory based trend lines for the currently selected Datacenter. Highlighting an area of the graph will give us some exact measurements for each workload at that point in time. We will now scroll back up to the top of the dashboard.

image

Once returning to the top of the dashboard, we can see that our clusters are out of balance. We would like to perform an automatic remediation to even the workloads between both clusters. We can do this by clicking on the "Rebalance" link and following the wizard.

image

The first part of the wizard gives us a graphical representation of what the CPU and memory workloads, along with storage utilization, will look like before and after the rebalance action happens. We can move to the next step by clicking "Next".

image

On the next screen, Step 2, we can see which VMs are going to be moved and where they will be moved to. In this example, we see the virtual machine named "appsvr01" will be moved from the Demo cluster to the DRDemo cluster. We are ready to start the rebalance and will click "Begin Action" to proceed.

image

We then receive a message indicating the rebalance action has began and lists a Task ID. We can reference this Task ID within vRealize Operations Manager's "Recent Tasks" area which is located in the "Administration" section. We can now click "OK" and wait a couple minutes for the rebalance to complete.

image

After a couple minutes, the "Workload Balance" dashboard will refresh and we can see how the rebalance action impacted our clusters. In this example, we can see the clusters' overall status moved to now be "Balanced". We can also scroll down the dashboard and see how the rebalance action impacted the individual resources as well.

image

Scrolling to the capacity utilization section, we can now see our resources are clustered much closer together. This concludes our walkthrough of VMware vRealize Operations Manager’s Workload Placement - Rebalance feature.

Filter Tags