We use UIM primarily for monitoring and managing all of our clients. As an MSP we need some kind of tool to be able to go out and do that. CA UIM has fit the bill quite nicely.
It's been pretty good. It does have its hiccups. It takes a fair amount of customization, but overall it's done what we need it to do and, when it doesn't out of the box, we are able to actually go in and tinker it enough to make it suit our needs quite nicely. So it's been doing a pretty good job.
It's the outcome of our being able to perform our job and gather the information we need in order to provide value to our customers. The real value is our being able to pull all the historic data that we need in order to gather every little metric and nuanced piece of information from a given device, a given piece of infrastructure, in order for us to generate alerts.
If we want to notify our clients, or for us to turn around and maybe run reports to proactively identify an issue with a device, that may not seem evident at first. But when you have all the pieces of the puzzle together, and when you can see these metrics aren't necessarily lining up nicely and they're acting a little bit abnormal, we can put that together for a bigger picture and provide a recommended action.
The convenience of both the web portal as well as the thick client. We are able to go in and actually leverage the thick client for a nice, easy drag and drop solution. It works out really nicely, especially for our front-end team. The team that I am on is actually quite small, so we need to leverage our front-end team to assist with making changes, and actually running the software. Having the GUI, that simple user interface, really helps us to offload that work to that front-end team to help free up our time.
Usually it's reliability issues. Some of the components don't always function as intended, or don't function as you would think they should.
To give a specific example, within this product there are individual probes, and each of these probes doesn't always necessarily output the same kind of information into our database. So when we try to collect QoS data, from one probe we might get a ton of information, lots of good stuff that we can use in our database, but then from another probe we might not get so much, or we might not be able to pull the things that we want to.
I'd also like to see more probes and more templates. More probes in the sense that we were coming across devices that we're expected to monitor and manage for which, out of the box, there isn't a nice clean solution. There are probes that are dedicated for certain devices and certain device types, which is great. But then there are times we come across nuanced products for which we have to develop our own solution. There are probes that exist in there that allow us to make a customized solution, but it takes a lot more time.
The other element is that there are no real templates, out of the box. Let's go with an example where we do have the probe, which is great, and we do have a really nuanced customer with a small set of devices that maybe not a lot of other customers use. There might not be a template in place, so effectively we have the tool in front of us but we still need to develop a solution. So it would be really nice to see a little bit more of something like a central repository of templates that we could use. That would help us expedite our onboarding process.
Stability is pretty good. As I said, it does have its hiccups at times. We use an on-prem solution, so for the most part it's pretty good.
Our issue comes in scaling. We're hitting a point with our customer base where we brought on enough clients in using the software that we're starting to hit some interesting connectivity hurdles. As a result of that, some things can take one or two attempts for it to work the way we want it to; maybe probes closing out or a connectivity from our primary infrastructure right down to one of our client sites. It can take two or three approaches, but overall its pretty good. There just are those occasional hiccups.
They're good. Their response time has been pretty good. They don't necessarily always have the answer I want, but that may be because the answer I want doesn't necessarily exist, so I can't fault them for that. But overall the response is pretty good. As far as the turnaround time, it's fast, the quality of the support answers that we get is usually pretty good.
There have been times where its been a little bit insufficient, in the sense that we might have needed to move it to a different engineer, or they may need to escalate it to go to their development team and get some answers, and the time frame on that can slow down a little bit. But for the most part, it's been pretty good.
I was not involved in the initial setup but we did a major refresher and I was involved with that.
It was pretty easy actually. It's a relatively simple project to set up. It's difficult to master but easy to initially deploy and configure, so I'd say it was pretty easy, off the bat.
My most important criteria when selecting a vendor are
Know what kind of products you are going to be using UIM to monitor and manage, and ensure that the compatibility is there. Because if there is not out-of-the-box compatibility, it can take a significant amount of time and effort to make things work.
Very valuable inputs regarding CA Unified Infra monitor ; Thank you for sharing Ravi Suvvari