VMware vCenter Operations Manager: An InfoWorld visual tour

VMware's smartly designed and extremely useful monitoring tool puts virtualization performance, health, and capacity planning at your fingertips

Intelligent monitoring for vSphere

VMware's vCenter Operations Manager allows administrators to peer deep into their virtualized infrastructure. Operations Manager continuously monitors each object, from VMs to hosts to storage, and collects a massive number of metrics regarding performance and availability. Based on that data, it then calculates normal operating levels for each VM, host, and cluster and determines where problems may exist.

Virtualized infrastructures contain a lot of moving parts, and vCenter Operations Manager is well suited to keeping track of all of them. Read the full InfoWorld review.

Dashboard view - World

Using a scoring and badge system, Operations Manager's dashboard views provide a clear picture of the infrastructure's Health, Risk, and Efficiency levels. Here we see a dashboard view of the "World" -- that is, every monitored vCenter Server instance. Operations Manager reports this environment is in good health but not very efficient, with 113 vCPUs, 280GB of RAM, and 690GB of disk space essentially going to waste.

Dashboard view – Cluster

In Operations Manager's dashboard views, you can drill down into each object (cluster, host, VM, and so on) to get the same Health, Risk, and Efficiency assessments at lower levels. Here we see a dashboard view of a single cluster, showing the same metrics as in the previous view, but now compiled specifically for the cluster, not the entire infrastructure. 

Operations view - Cluster

For any monitored object, click the Operations tab for a display of key metrics and overall health. In this Operations view of a cluster, for example, we can see a chart of overall health at the upper left and graphs of individual metrics (CPU, memory, network I/O, and disk I/O) below. On the right, we see a graph of abnormal metrics over time.

Operations view - VM

Here we see an Operations view of a VM, detailing the overall workload, CPU, memory, and I/O metrics. The brackets next to or above the workload bars show the normal range of operations. It's easy to see that CPU, memory, and network I/O levels are fine, but disk I/O is far greater than normal.

Analysis view - Cluster

The Analysis tab in vCenter Operations Manager helps you zero in on hosts, VMs, and other objects based on performance characteristics. For example, this Analysis display for a cluster reveals which VMs have a high workload that may be due to long-term stress. You can just as quickly discover which hosts have the most free capacity and least stress, or which VMs have the highest CPU demand and contention, and so on, with the resulting heat map and detail below.

Planning for growth

By moving through the tabs you can quickly find where current or forecasted problems lie, and you can work through active alarms. In addition, you can use the Planning tool to create different what-if scenarios that can greatly assist in planning for the addition, removal, or resource restructuring of current physical hosts. Here, under the Planning tab, the Summary view shows that at the current rate of resource consumption across the environment, current capacity will be sufficient for at least a year.

Avoiding false alarms

Operations Manager mines historical data on every object in the virtual infrastructure to avoid false alarms, triggering alerts only when activity veers outside the bounds of established norms -- even taking into account time of day and day of the week. Here, in the alerts display for a cluster, we can see the graph of alerts over time, and the list of specific alerts at the bottom.

Predefined reports

Operations Manager provides a number of predefined reports that can be scheduled or run on demand. These include reports on undersized or oversized VMs, idle VMs, host utilization, capacity overviews, and so on. Note, however, that you cannot generate custom reports from within the GUI.