What is most valuable?
It gives us insight into the functional health of the application, if we have any hot spots. For distributed applications we need those insights. We like the dashboards where we can consolidate different metrics into one view. Also, it tracks back-end service calls and it gives us insight into any issues with the services we are calling, as we use distributed application technology. Also, alerts and notifications.
How has it helped my organization?
The organization becomes quick in assessing hot spots and attacking any issues that are bringing down and application. It means meeting SLAs and reacting to issues quickly before they become problems when we see patterns, before the whole system becomes unusable.
What needs improvement?
It is a technology with a lot of things going on. We'd like some use cases to help us configure APM for dashboards and reporting.
Migrations from one version to the next are always a battle and we have to involve architects and other consultants to help us with upgrades, which is expensive.
It’s enabling technology, so we shouldn’t be spending so much time and resources managing the application.
What do I think about the stability of the solution?
We had some hiccups, and still are. When we implemented agents for one of our applications, the agents had a memory leak and halted the application, so it was unstable. We were told to upgrade to a new version of APM which mostly addressed the issue.
The issues are ongoing to an extent, and we’ve been working with CA for a while on a trial and error basis, which has been painful.
What do I think about the scalability of the solution?
It is scalable, but we have to go through a lot of pain in configuring agents. Agent configuration has been something of a mystery because of poor documentation. We talk to CA and we get some guidance, which often doesn’t work. When you’re rolling out agents 4000 JVMs, it is a painful process, and that’s only one application.
How is customer service and technical support?
We don’t have dedicated resources, but we open tickets and they come back. For example we’ve seen a lot of errors in our logs, and they’ve provided patches and configuration changes, but we’re still seeing those errors. It’s not bringing down the system but we don’t know what lurks, and we want to make sure that it’s stable. Sometimes we go to our account manager to escalate, which does get us an answer.
Still this has been a nagging pain overall, although the solution was more solid after upgrading to newer versions. We do seven to 10 billion dollars in revenue in those applications so we need to ensure those systems are as available as possible, and that hasn’t happened.
How was the initial setup?
It was already in production when I joined.
What other advice do I have?
It’s been a battle with upgrades, stability, and, to an extent, support. We also gain some from it – I will say it it’s been very helpful for us to identify issues but it could be better. The industry is evolving and there are pretty good cloud-based tools out there that offer me more than the current version of CA APM offers. Based on prior experience, we are dreading the upgrade to the next release. In a cloud environment, we don’t have to upgrade or manage servers.