Home > Blogs > VMware Consulting Blog


Create a One-Click IT Command Center Operations Dashboard using vRealize Operations for Horizon

Anand VaneswaranBy Anand Vaneswaran

In my previous post, I examined creating a custom dashboard in vRealize Operations for Horizon that displayed my current cluster capacity metrics in my virtual desktop infrastructure (VDI) environment. This helped provide insight into the current utilization and performance. In the final post of this three-part blog series, I’ve provided instructions on creating a one-click IT command center operations dashboard. Many enterprises tend to centralize their IT command center operations in an effort to coordinate multiple technology focus areas such as network, storage, Microsoft Exchange, etc. and bring them together under one roof. The idea is to be able to see, respond to, and resolve incidents that cause production environment outages that have a wide range of implications. It is also to increase efficiency to speed up response times, and create a centralized view of the overall environment. In a production VDI environment, the onus would then fall on the command center to be able to incorporate VDI as a technology focus area. In this blog I’ll explain how to create a one-click dashboard to focus on certain key stats that are central to the VMware View environment, and help the command center personnel in times of outages.

As I have stated in previous posts, these are examples that can either be replicated in their entirety, or be used as a jumping-off point in an effort to construct a custom dashboard with stats that are most germane to your environment and command center personnel.

Additionally, as I have done in previous posts, I’m going to rely on a combination of “heat map” and “generic scoreboard” widgets. I’m also going to introduce a widget type known as “resources” in this dashboard. In total there should be nine widgets:

  • Four generic scoreboard widgets
  • Three heat map widgets
  • Two resources widgets

The final output should look like this:

AVaneswaran 1

I then want to configure my widgets so the following details are presented:

Heat maps

  • The overall health of my ESXi hosts running full clone VDI workloads
  • The overall health of my ESXi hosts running linked clone VDI workloads
  • The overall health of my Horizon View infrastructure servers. These servers include my Connection Servers, Security Servers, vCenter Server, View Composer server, etc.
  • The number of available virtual machines for linked clone pools. This is an important stat for considering the environment, so you can find out the maximum number of desktops in the linked clone pools.

Generic scoreboard widgets

  • I want to check to see if my Connection Servers are enabled and accepting incoming connections. This particular stat will take on added significance in View environments running a 5+2 Connection Servers replicated ring.
  • The number of concurrent external connections currently accessing the environment.
  • And finally, a multi-purpose widget to provide the following data:
    • The total number of connected concurrent sessions.
    • The total number of overall virtual machines in the infrastructure.
    • The average bandwidth utilization per session. Horizon View desktops can experience anywhere from 150 Kbps–350 Kbps for task and knowledge workers utilizing apps such as browsers, Microsoft Office, and basic productivity apps. This figure increases with high graphics, printing, peripheral device, and audio and video usage. However, if I’m confident the environment is running a fairly uniform set of workloads, this stat is useful for monitoring the entire environment. If the pools are built based on varying use cases with different workload profiles, it might be a good idea to generate this stat on a per-pool or per- use case basis.
    • Outgoing and incoming packet loss on the network vLAN segment running my VDI workloads.
    • Total network utilization on the vLAN segment running VDI workloads.
    • Total bandwidth utilization on the network vLAN segment running VDI workloads.

Resources widgets

  • This widget calculates the average PCoIP round-trip latency in milliseconds. This is one of the most important and oft-monitored stats in a Horizon View environment with centralized infrastructure that also serves end-users accessing the environment from distant locations across the WAN. I want to achieve this result on a per-pool basis, and am only concerned about external connections coming in from WAN locations. In addition, I’m particularly interested in low-bandwidth, high-latency remote sites, but not so much with internal users connected to the corporate network with regulated and guaranteed bandwidth. Finally, I want to configure this first widget for my automated full clone pools.
  • Next, we replicate the aforementioned widget, only this time for automated, linked clone pools.

The widgets can be arranged in the dashboard however you choose.

To start, I’m going to configure a full clone widget to display the health of my ESXi hosts running full clone desktops, and I want to place the heat map widget on the far left side of a three-column dashboard.

AVaneswaran 2

The key here is to filter by the hosts running full clone desktop workloads. This is achievable with a custom resource tag I’ve created in my environment. I’ve demonstrated the technique to create such a resource tag in the first of this three-part blog series. The configured widget should look like this:

AVaneswaran 3

Repeat this procedure for another widget for linked clone desktop pools, and filter by the hosts running linked clone workloads.

AVaneswaran 4

The configured widget will look like this:

AVaneswaran 5

Next, I want to configure the following widget to display the health of my View infrastructure servers, and I want to place this in between the first two widgets along the top portion of the dashboard. It is important to place the infrastructure server resources in custom resource tags so they are filtered by said resources.

AVaneswaran 6

Here is the configured widget:

AVaneswaran 7

Next, a generic scoreboard widget is placed underneath the heat map widget we just configured. This widget will display the number of enabled connection servers that accept incoming connections.

AVaneswaran 8

When complete it will look like this:

AVaneswaran 9

The next step is a generic dashboard that displays just the total number of tunneled sessions through the View Security Server.

AVaneswaran 10

And here is the end result:

AVaneswaran 11

We now want a heat map that displays the number of available virtual machines in the automated linked clone pools. In order to ensure production pools are more or less consistent during peak times, we need a heat map that shows the maximum number of desktops and total sessions.

AVaneswaran 12

Once again, the trick is to filter by a resource tag for your automated linked clone pools; the heat map will look like this:

AVaneswaran 13

Next I want to work on a generic scoreboard that gives me the following details:

–        Total number of current concurrent sessions

–        Total number of overall virtual machines in my environment

–        Workload percentage on the DHCP vLAN, which is serving all VDI desktops IPs

–        A super metric that calculates the average bandwidth utilization in Kbps

–        Outbound and inbound DCHP vLAN packet errors

The widget should be configured like this:

AVaneswaran 14

Super metrics are required to calculate the average bandwidth utilization per session, and the total DHCP vLAN bandwidth utilization. Here is the super metric for calculating the average bandwidth utilization per session.

AVaneswaran 15

We also need a super metric to calculate desktop DHCP vLAN total bandwidth utilization.

AVaneswaran 16

Finally, configure the two resources widgets. The first widget goes on the bottom left of the dashboard, and is configured as follows:

AVaneswaran 17

The end result will appear like this:

AVaneswaran 18

Make sure to filter by the custom resource tag containing only full clone pools. Replicate this process, step-by-step, on the bottom right-hand side of the dashboard, but this time for linked clone pools.

And here is the final dashboard!

AVaneswaran 19

In conclusion, here are a few takeaways from this blog:

  • IT command centers are sometimes challenged with knowing exactly who to assign an issue to, and typically can use more visibility than they can get. In providing this custom dashboard through AD Security Group dashboard sharing, you can help your command center personnel get exactly the kind of visibility into the environment they need to aid them in their decision-making process.
  • The examples provided in this three-part blog series show you the extent to what you can achieve in vRealize Operations for Horizon in a time-efficient manner. Fundamentally, it’s a matter of knowing what data you want to display, and if done correctly―as demonstrated in these blogs―data manipulation becomes extremely easy.

Now, I’ve barely scratched the surface of VMware vRealize Operations Manager capabilities in these blog posts; there is so much more that has not yet been discussed. I just wanted to focus on a set of custom dashboards, where each one is designed to achieve a very specific purpose. The methods detailed in these blog posts only demonstrate one approach – but there are others. These show just some of the ways vRealize Operations Manager can be explored, data can be mined, and ways you can gain a view into the environment.


Anand Vaneswaran is a Senior Technology Consultant with the End User Computing group at VMware. He is an expert in VMware Horizon (with View), VMware ThinApp, VMware vCenter Operations Manager, VMware vCenter Operations Manager for Horizon, and VMware Horizon Workspace. Outside of technology, his hobbies include filmmaking, sports, and traveling.

Leave a Reply

Your email address will not be published. Required fields are marked *

*