One, it's agent-based. It allows us to do multiple things at once as opposed to serially, which we had a home grown tool that did it serially. We've reduced our time frames. I think the pre-codified actions for doing the release activities, also is a major benefit. It's something that we don't have to write scripts for, they're already written. We just need to pull them together. Those are the two major things.
The other thing that the tool does that we're beginning to realize is it's addressing what Gartner's calling this bi-modal approach. What we're realizing is that Mode 1, which is your old legacy type of approach, which is why we got the tool. It's fantastic for, but now the new fast speeds using it in the Mode 2 type thing, we're finding it bullet-proof there as well.
Improvements to My Organization
One of the key benefits, what really we're looking at is the quality of the releases, not necessarily the code quality, but the actual deployments and the speed at which we're doing it. Things that we automate are taking upwards of 8 to 10 times less. Now, what we can do is we can do 3 to 4 times more on a release. We just had our largest release ever. Without CA Release Automation, there's no way we could deliver that level of code to our end-customers. It's very stable.
Room for Improvement
We're using it more on the deployment side. What we really want to do is see a stronger or easier user interface. I think the biggest complaint that we've gotten from the development side is that the user interface, the experience is not as good as they'd like it to be. It's hard to comprehend. We've overcome that with a lot of our own solutions, but we'd really like to see that user interface, and would like to see them get into more of the release orchestration aspect. We're actually using an internal solution for that. I know they're heading in that direction, we're really looking for that, and again, that needs to be ease of use. That user experience is critical. The developers have their critical time frames, and for them to figure out and comprehend a new tool is not acceptable. We need it to be simplistic and easy to use. There are some challenges, we've overcome a lot of them though.
Very stable. We were fortunate that early on, one of our internal technicians/architects sat down with CA. Went through the process of designing what the solution should look like. We're benefiting from that up front set up and structure of the tool in order to make it very reliable. It's enterprise wide, it's being used by all groups.
Extremely scalable, that's one of the main reasons why we bought it. It's extremely scalable. We've been able to scale up from probably 100 or so automations to where we're doing 700 now, per release. It's been a very powerful tool. Broad set of technologies as well.
Customer Service and Technical Support
The tool has been so bullet-proof, we haven't had to tap it, but when we have tapped it, it's been spot on.
Our releases were getting bigger and bigger. The demand was getting larger and larger. We knew that the home grown tool was not going to be able to be scalable, basically. It was elongating our releases. We need to get our stuff done in a weekend, and it was beginning to drift into the following week. We went to the marketplace, CA Release Automation was a market leader, we made a quick decision in that direction. We haven't regretted it.
The support we've had with CA in the past had been very strong, those were important to us. Then, the depth of the solution. I think we're looking into the depth of the solution, and the agility and flexibility of that solution, I think were important to us in this decision. We're with a large player, somebody that could support us if we needed. The tool, from everything we had been reading and researching, it could stand up to it. I think those were the things that we were looking for.
The initial set up itself as relatively straightforward, what was a challenge was the roll out. You had competing priorities, you had adoption issues, you had skill set gaps. Not only on the roll out team, but on our development team. What we were able to do was close some of those skill set gaps. It became a journey where we had to show the value to each of the individual development groups, and it created momentum. As we got these development groups onto the CA release automation, they saw the benefit. We were able to take internal environments, and be able to turnaround those environments in seconds as opposed to days. We began to get the adoption. It's getting that initial movement forward which was really the challenge. I don't know that there's anything that CA could've done. It was something that we had to do internally.
Other Solutions Considered
At the time we did the selection, no. They were out there. I was not part of the selection process, I know the person that was, and we got release automation before it was known as CA Release Automation. It was then purchased by CA. It was a clear market leader, so we basically went in that direction.
My recommendation would be: a) buy the tool, b) recognize that it's a journey. It's not something that it's a binary thing that's going to happen overnight. You've got to continually sell it and treat it as an ongoing initiative. We're post the projectization of it, and we're still bringing people on board to the process. I would recognize that this is a continuous delivery, if you will, exercise that I don't think ever stops because your code and your code base is constantly evolving. There are things that you can continually do, integrate it with your change management environments, integrate it with some of the other DevOps tools that you can't even imagine what you can do with it. I would say, recognize that it is a great foundation for your release automation, but it doesn't stop there. This tool is holding up to the new requirements that are coming, and that's what we really like about the tool is as the world evolves, this tool is stable. It's stable and it's able to adapt.
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Jan 20 2016