Introducing Login VUM Beta: User Experience Monitoring Solution for VDI
Hey, Adam here, Product Manager at Login VSI. This week at VMworld we’re announcing a new product beta. It’s something I’ve been working on since joining the Login VSI team earlier this year, and I’m excited to share the news with you. We’ve been calling the product Login VUM throughout development.
Login VUM: What is it?
Back in the day, coal miners used to take canaries into the mine with them. Canaries are more sensitive to bad air, so when the canaries started to get sick or faint, the miners knew it was time for them to get out before the toxic air (usually carbon monoxide) affected them. The Police even wrote a song about it. VUM is like a canary for your VDI and RDS environments – letting you know when things are getting unhealthy for your real users. This is a big deal for customers – lots of attendees at events this year told me they don’t find out about issues with VDI until users start calling them to complain. The other tools that are out there are complex and expensive, and none are as effective at giving insight into what the user experience is like. A system could be reporting 100% CPU capacity, but still be delivering apps just fine. VDI is funny like that. VUM will tell you what your users are seeing, regardless of what your expensive management consoles say.
How does it work?
VUM is based on Login VSI, but instead of simulating a full user workload with multiple users to test the capacity of a system, it simulates just one user on a production system. A virtual user simulates a regular user on the system, spending all day just logging in and starting apps, and timing how long it takes them to complete. When it notices a problem – like suddenly not being able to login, or apps taking much longer than usual to start – it alerts an administrator to the issue.
Nope, there’s more. There’s a web console that shows you the results the virtual user is getting, along with showing you the alerts and performing management tasks:
There are a lot of VDI monitoring tools on the market already – how is this one different? Tools tend to be categorized as either “user-experience” or “holistic” monitoring tools. I thought it would be useful to describe the differences between the two, and explain where VUM fits. A holistic monitoring tool looks at all the components of a system and reports their utilization. It typically has many charts or graphs showing CPU utilization, RAM utilization, bandwidth consumption and more. Like choosing car from the list of specs, there’s a lot of data, and you need to decide which data is relevant. On the other hand, user experience measurements typically boil everything down to one measurement: Do users find their desktop performs acceptably? Login VSI uses the VSImax score to measure this – it’s the number of users that can use the system before performance is too slow/unacceptable. It’s like comparing just the top speeds of cars to decide on one.
Holistic solutions are hard to build. They require knowledge of the underpinnings of several different components, and they’re all different depending on which hardware and software solution you build. They are also sometimes difficult to interpret – just like how horsepower doesn’t always mean a faster car, high CPU doesn’t always indicate a problem.
VUM is a user experience monitoring solution. We decided on this approach because we already have an engine for measuring user experience in Login VSI, so we could build on that to build a production monitoring system.
And here’s the thing – long before they started doing VDI, IT shops have had other servers they needed to monitor, so they almost always have some sort of holistic monitoring system already – it’s just more generic and not specific to VDI. If VUM can let them know there’s poor user experience happening, they can look at their existing monitoring tools to figure out exactly where the slowdown is coming from. Deploying a whole dedicated holistic VDI monitoring system is probably redundant and overly complex. Our goal was to build a companion solution that can be easily integrated into existing systems, or stand alone, to provide insight into how users feel the system is running. We’re building connectors that will drop the VUM data into the other operations tools that you probably already have.
We’re currently running an invite only, private beta, but will be expanding that to a public beta soon – watch this spot for info on how you can participate.