What is our primary use case?
The key use cases will be eliminating manual activities, reducing the risk of missing steps in the deployment. The other thing is to speed up deployment, because the previous way of working was having someone to document the steps, someone to review the steps, and then at the end of the day, someone had to execute those steps within that window.
So far, it has performed well, at least from the "repeatable" perspective. All deployments are repeatable, so when we have issues with deployment, it gives us the opportunity to review and to know where the issues came from. If it is working now and was not working previously, we usually know where it fails.
How has it helped my organization?
I think on a day-to-day basis, it has increased the capacity to deploy. We don't have to wait for someone to do something. As long as it changes, then we can always deploy to an environment on demand.
What is most valuable?
I would say the traceability part of it. With this feature, I know which environment is running what. Which version of the binaries; that is key because then we know what to fix.
What needs improvement?
I think I spoke about this to them. The key thing is support for cloud-based deployment. That is lacking. Today, the whole world is looking at cloud deployment, running a cloud application. But it doesn't seem this platform will have that feature any time soon.
What do I think about the stability of the solution?
Stability-wise I think it still needs to be improved; performance-wise. It crashes from time to time. Sometimes it just hangs and requires a restart. Upgrades, depending on the versions, can be tedious and risky. We recently had a problem with one of the version upgrades, and the platform was down for a day. That wasn't a very pleasant experience.
What do I think about the scalability of the solution?
I think scalability-wise it is proven. We are running a significant number of end points. I think other customers have run larger number of nodes as well. The scalability should be okay.
How is customer service and technical support?
I use it all the time. The level of response is very specific to the individual that answers the call. Some are knowledgeable. But there are times that we are left on our own. The response time itself, overall is okay.
How was the initial setup?
Initial setup is definitely easy. But when it comes to growing, the system becomes a bit more complicated. You want to have that HA capability, then you have to have redundancy in terms of connecting the different nodes, and you have to test that. Sometimes nodes don't seem to talk to each other. That's a problem.
What other advice do I have?
At this point in time we are not investing more because we already bought upfront. We are taking a wait-and-see attitude because, if the maturity of the platform plateaus, meaning we don't have new features, we might decide to move on.
When we are looking for a new vendor, what is important to us is ease of migration. The other factor is the the currency of the platform, how current is it in terms of alignment to the market we are in.
I give it an eight out of 10 because it does about 60-70 percent of the work. What it does, if it's done correctly, it does well.
I would advise a colleague who is considering this type of solution to look out for the fact it doesn't support the new stuff. But if you're looking for solutions that are based on your existing, traditional infrastructure, I think that's good tool.
Disclosure: I am a real user, and this review is based on my own experience and opinions.
Dec 07 2017