This post is republished on our blog with permission from Rob Beekmans, vExpert, CCEE, CCIA, VCP5, VCP5-DT & VCAP-DTD. The original post appears here on Rob’s blog, EUC, UEM, Mobility and Monitoring.
In this article I will discuss why I think it is important for vROps to become a more open monitoring platform so I can use it to monitor my entire environment including non-VMware infrastructure to proactively improve the end user experience.
I love vROps and want to do more with it. This blog describes one way to achieve that goal.
Useful Data in a Single Console
Important in monitoring, those who read my articles know this by now, is to get the right information and preferably from one console. You don’t want too many consoles where you, as an admin, are combining info from all of them leading to false positives or worse.
What do you want to see when you look for a monitoring product?
- Visibility in some kind of layering (Hypervisor, VM’s, OS, Applications, physical server);
- Monitor all products from one console (VDI, RDSH, storage, network etc.);
- Monitor all hypervisors from one console;
- Send metrics to current monitoring products or receive metrics you need.
Ideally you monitor with one product only, I haven’t seen that yet but it would be utopia.
Many Enterprises have monitoring in place – CA, HP OpenView, BMC, IBM or anyone I forgot here. Optimally you would want to feed these products with information, again to add to the central console idea.
These enterprise products, however, allow other monitoring solutions to fill these “gaps” by allowing information to be pulled through API’S or using syslog or SNMP traps. This is really important because no enterprise monitoring solution covers everything equally well.
Monitoring Everything Outside of the VMware Stack
I think vROps can do the same thing and I think they should. You could use vROps to monitor everything outside of the VMware stack including Citrix XenApp, XenDesktop, network elements, applications and even physical IT infrastructure elements. VMware is almost there. Think about it.
VMware already has a clean interface that could easily suit as the central interface to monitor your environment. From a vendor monitoring solution, vROps is one of the most complete solutions on the market, it monitors many aspects of the virtual environment and you can add management packs and extensions to extend it.
It is focused on resource management, giving insight into how your desktops are performing. Many default dashboards and reports complete the product. So far it’s all good.
Two Key Issues with vROps: EUE & E2E
VMware vROps has two issues in that it doesn’t understand the environment from an EUC perspective and it doesn’t do E2E monitoring. Both “issues” are close neighbors, for one would help the other and they relate to monitoring limitations I mentioned above since vROps doesn’t monitor anything beyond the VMware stack.
First, if we look at E2E you need to monitor outside the VMware stack and you need to take into account that e.g. DNS is a factor in the service delivery of the desktop. Not many products have that ability, most products monitor a specific set of components. vROps is no exception.
VMware could create a dashboard where the Citrix monitoring can be combined with the other management packs so that you could create your own E2E dashboard. Seems pretty easy to me. If we look at EUC we see something else.
Second, if we look at EUC we see something else. vROps monitors from a VMware point of view leaving out important metrics that really show the user performance. You can’t measure user experience from looking only at resources! The only way to measure the user experience is to measure what the user is actually doing. Simulated user logon this is called.
Measuring with simulated user logon would solve a part of the missing E2E part. If you work like a user you will test a lot of components that make up E2E.
Simulated User Logons & Real User Experience
Simulated user logons would show the real user experience, add that to vROps and the overall color of the VDI environment would really show the performance. Of course, this is pretty hard to get into the product for it would need to know what is added and take that into account. They could add a new option in the dashboard to show metrics related to user experience.
Metrics like:
- Logon Failures
- Logon Failure Alerts
- Logon time
- GPO load time
- Profile Load time
- Break down of connection steps (broker, starting client, etc.)
- Root cause analysis.
VMware has no option to do this. There is only one product in the market, Goliath Technologies Logon Simulator, that both simulates logons, exercises the whole desktop delivery infrastructure and can also tell you where the root cause of any issue that impacts EUC experience because it is truly end to end. The question is why can’t they deliver this data to vROps?
There are other products like Horizon View logon and LoginPI that come to mind instantly. The issue with these is simply that they don’t exercise the whole virtual desktop delivery infrastructure and, because they are not a feature in a full enterprise monitoring product, they can’t determine the root cause of the issue.
Integration that Makes Your Life Easier
I’m going back to my last articles about Goliath, they integrate with VMware vSphere vCenter so that you can combine information gathered from both products. This makes managing an environment so much more complete.
I included some screenshots I borrowed from Goliath to show to data you can gather that makes your life easier. I could have taken these screenshots also from other monitoring vendors for they also add value to VMware vSphere vCenter monitoring.
The point is that with integration you add value, suddenly they are not two products running side by side, they are products making a suite adding value for you.
A Solution to Gain Visibility into the End User Experience
With VMware vROps being closed, or so it seems, when you have vROps running you will need another monitoring product to monitor your virtual desktop environment to get the user experience. This to me seems like something that could be improved.
My idea with this would be to add a segment in the VDI dashboard that would allow external parties to add metrics. It would not hinder the current dashboards and metrics but would get a nice place in the dashboard for customers to add if they want to. I talk to many customers about this, user experience monitoring and end to end visibility, and that’s what they are looking for.
So VMware could you please, pretty please open up vRealize Operations Manager to allow other monitoring vendors to add more value to your product? You can name it Project V8 – Putting power in your vROps.
Lastly, I encourage you to watch an on-demand version of a live webinar focused on improving the VMware end user computing experience by gaining visibility into Application & OS performance data in addition to VM, Host, and Hardware to proactively remediate issues before end users are effected.