By: Kyle Wassink, Blue Medora
Monitoring, understanding, and optimizing compute performance is a crucial aspect of any IT environment, and it becomes even more important with virtualization. VMware vRealize Operations (vROps) helps you realize the benefits of virtualization without sacrificing visibility into the compute layer. vROps also makes optimizations easy with performance trend analysis, health badges, project planning, and more.
Prior to virtualization, a server might have hosted a database, a few applications, or simply have been a development server for testing against. In a virtualized environment, however, it’s likely that there are far more databases, applications, and test Virtual Machines relying on that same piece of hardware. Increased server consolidation and Virtual Machine density means more business-critical applications are relying on a single compute node for uptime and physical resources.
VMware Host: Default Visibility
The importance of the compute layer is readily apparent in vROps by the extensive functionality available for Host Systems (VMware Hypervisors). Everything from thorough badge integration to capacity planning to views and reports is provided out-of-the-box. Of course, you can also create new views, reports, dashboards, alerts, and more to leverage even more insight into your VMware Hosts. Below is an example of the integration with the Capacity Remaining badge which provides analysis on host resources such as CPU, memory, and disk.
Figure 1: Capacity Remaining badge
The level of detail and analysis on host systems in vROps provides a phenomenal starting point for understanding the compute layer in your environment, but you can do even better. Vendor specific management packs are available on VMware’s Solution Exchange to dig even deeper into the compute layer.
With so much functionality provided out of the box in vROps for hosts, you may be wondering what the value is in additional compute layer information. The answer is simple: specialized insights. The additional insights provided by compute layer management packs help make sense the hardware underlying the VMware host and avoid problems through early detection. For example, compute layer packs usually provide relationship mapping to the VMware hosts, thus linking the vendors compute topology to the virtualization topology.
Figure 2: Dell PowerEdge Traversal Spec
Cisco UCS has recently claimed the largest market share of x86 servers in the United States. Worldwide, they rank second for the same measure. UCS is everywhere, and chances are good that you are running some in your environment. If you are, you can monitor it in vROps with the Management Pack for Cisco UCS.
This Management Pack provides over 350 UCS-specific alerts in vROps, covering power redundancy failures to firmware problems to temperature tolerance violations. All of this is UCS-specific, meaning it is not otherwise available inside vROps. Wouldn’t it be nice to know if your redundant power supply just failed on a mission critical VMware Host? With the Management Pack for Cisco UCS, you can see this right in vROps. Most of the alerts even provide Cisco’s recommended steps to resolution inside vROps!
Figure 3: Cisco UCS Alert Menu
Another great aspect of this pack is the addition and integration of the UCS stack, from Chassis to Blade to Fabric Interconnect and more. A relationship is automatically created between the UCS Blades and the VMware Hosts as well, providing an easy way to jump from one stack to the next.
With the largest market share worldwide, HP Server customers can also take advantage of a specialized vROps management pack. The Management Pack for HP Servers provides vROps visibility into HP chassis, blades, racks, network adapters, and more. Relationships are created between HP servers and VMware Hosts, allowing easy identification of Host and VM dependencies on HP hardware.
The Management Pack for HP Servers also provides custom dashboards that help identify problems between the HP and VMware stack. For example, the HP Hosted VMs dashboard lets you select a VM and view the VMware Host, the underlying HP Hardware components, and KPI metrics at each layer. Alerts are also collected from the HP environment, notifying you right within vROps when an HP network adapter goes down, if there is a physical memory problem, and more.
FIgure 4: HP out-of-the-box dashboard content
Dell PowerEdge servers also make the top 5 list of most popular servers worldwide. For customers running PowerEdge, vROps integration is also available with the Management Pack for Dell PowerEdge.
With this Management Pack installed you’ll get Dell components integrated into vROps, complete with relationships to the VMware Hosts they support. Alerts and events from Dell are captured and displayed in vROps, alerting you when server cooling and power components fail, server storage has problems, and more.
FIgure 5: Dell PowerEdge Traversal Spec
Past the compute layer
vRealize Operations has a vibrant and comprehensive ecosystem of management packs, covering nearly every aspect of your IT stack. The three compute packs covered today are just a sampling of what is available. Be sure to check out Blue Medora’s product page to find these management packs and more!