What is our primary use case?
Monitoring the servers, the infrastructure and we also monitor applications with a specific probe doing Synthetic transactions. We use a dirscan probe to monitor files to make sure that they transfer at the correct times, and it will send alerts if they don't. We use logmon monitoring and we use the event log monitoring processes. We monitor processes for up/down state, CPU usage, memory usage. We use the NT Services probe, the monitor services on the Windows boxes. That's to name a few.
Regarding performance, we've had some struggles with it at times, but we get a support case opened up and support has been very good at helping us resolve the issues that we encounter.
What is most valuable?
The monitoring of the applications to let our business know when things are performing and that they're up and available.
How has it helped my organization?
We have less downtime. We have alerting to let us know when disks are filling up so that we get that taken care of before it becomes an issue and is noticeable to customers.
What needs improvement?
In the UMP, certain devices will show up multiple times and they don't correlate correctly. That's one of the issues.
Sometimes the probe, on its first release, we will find some bugs with it and notify support and then they escalate it to the upper level and they get things corrected.
The dashboarding. They're going in the right direction, getting away from flash and using the HTML5 with the Cabbie dashboards. That has been very helpful with us in developing dashboards. But maybe some additional out-of-the-box dashboards with different standard tools that people are using.
The one thing that our company has started to use is MarkLogic, and they don't have a specialized probe for that. We've reached out to them and put feedback on the community trying to get votes on that. But so far, it hasn't gotten a lot of votes.
What do I think about the stability of the solution?
The stability is good. They have an HA feature, high availability. And whe we were setting it up with use of support, we decided not to even set up that functionality because it's very seldom that we have a problem with it going down.
What do I think about the scalability of the solution?
Scalability has been fine. As the new servers are brought on with the new MCS tool, it allows us to get configuration on the servers put on in a faster time.
How are customer service and technical support?
I'm happy with the technical support we've received, and their response time.
Which solution did I use previously and why did I switch?
We actually had UIM before it was UIM. It was Nimbus in 2004. Then it went to Nimsoft, then CA bought it and then rebranded it CA UIM. When it got brought into our company, in 2004, we used it for Synthetic transactions, to monitor the email and different products on the web, and the response times to that. And then we were using a different product for our network monitoring.
We wanted to try to eliminate some of the excessive tools we had so we moved our network monitoring into CA UIM at that time.
When we first moved on to UIM, and brought the network monitoring in, at that time, the event correlation product wasn't built in - so one event happens and then it triggers three or four other things. And when we were doing that, the product we were getting rid of did do that. CA had said that that would be on the roadmap. It seems like the roadmap has changed now, and they're doing more of the event correlation with Spectrum, but we don't own Spectrum. So we have a little bit of a struggle there with the event correlation, and it seems like CA is not doing the event correlation with their SNMP Collector probe. They've moved more towards Spectrum.
How was the initial setup?
It was straightforward. I and another person set up how the servers are going to be set up and then we got approval through CA. We asked them if that looked good to them, and they came back and with what we had set up, they didn't have anymore recommendations. They thought that what we're going to do was going to be successful.
What other advice do I have?
When our company is looking to invest in a vendor, our criterion is that we will try to stay with a vendor that we have a relationship with already.
I rate it an eight out of 10 because the ability to configure the probes is much easier than with other products. Before we went with the UIM product, I had to evaluate other products and the configuration of those was much more difficult than with UIM.
I would advise, because they have the new SaaS product - and I have a feeling we're going to be looking at that at our company also - doing a demo of the SaaS product and see if that meets their needs.