I'm not the kind of person who gets out of bed in the morning all jazzed up about monitoring products.
Everyone needs some product for monitoring, but when people ask which one I like best, I don't really have an answer. Agentless monitoring is great for its simplicity, but it's easy to fall in love with an agent-based monitoring tool because of the breadth of information you can get.
Dashboards and the like are cool, but if it's only reformatting what I can get out of PerfMon, I'm less interested. Besides, who spends time staring at dashboards? We have work to do!
Each product works differently, too. Some focus more on network, others on the OS. And there are monitoring tools for applications, the remote desktop protocol and the client side. You need all of these things to have a comprehensive monitoring strategy, and because of that, you probably use many tools instead of one that tries to do it all.
But all of that's in-house. What about desktop as a service (DaaS) monitoring?
Where are all the DaaS monitoring tools?
With DaaS, we rely on the provider to ensure our desktops are delivered to our liking, so we don't necessarily care about keeping tabs on the affect that applications, OS configurations and the network have on that desktop delivery -- that's what we're paying to not have to worry about.
Then again, what do you do if your needs aren't being met? How do you prove that your provider isn't meeting the terms of their service-level agreement (SLA)? Or, if the desktop is technically available but isn't delivering the right user experience, how can you pinpoint that?
I've yet to see any DaaS-specific monitoring tools, which is surprising given all the weight that's been put behind DaaS this year. There are SLA monitoring products that focus on software as a service, but DaaS doesn't seem to be getting any attention. Is there an opening for a product, is the market too nascent, or is it not worth the cost?
DaaS providers have SLAs, but when you look at what they're on the hook for in the event of an outage, you might be surprised. Amazon Web Services (AWS), for instance, only reimburses you for the time of the outage. So if you paid $35 per user, per month and were down for two hours, you're getting a fat credit on your next bill for 9 cents. Even if you have 2000 desktops on Amazon WorkSpaces and they're all down for two hours, your bill is credited just $180.
That probably doesn't happen every month -- if it does, you have bigger issues than finger-pointing -- but let's say it did. That's $2,160 per year of credits for not meeting the SLA. What would an SLA monitoring tool cost? Certainly more than that!
Even if one was available to you from the provider itself for 50 cents per user per month, you're still 41 cents in the hole.
Of course, this is all excluding the value of the downtime. If there is an issue, you need to be able to point the finger back to the provider and say, "You're affecting my bottom line in this specific way." The thing is that, DaaS providers already know that. If you were reimbursed for the actual value of your downtime, that throws off their business model -- and good luck calculating the value of lost time.
So perhaps we're not seeing DaaS monitoring because it just doesn't matter. It's one of the challenges of moving to DaaS, after all: You trust the provider to ensure everything is running smoothly. Without that trust, you shouldn't use that provider.
If there's an issue in an on-premises environment, you bear the burden of responsibility. The cost of the downtime is directly attributable to you and your systems. You absolutely have to have something to monitor them. But with DaaS, you're either meeting your SLA or you're not, and you have to be OK with the ramifications of that.
Monitoring, testing and tools help avoid VDI performance issues
How VDI monitoring tools straighten out virtual desktop snags
Four VDI monitoring tools to review