Stonebranch Universal Automation Center Review

Gives us better visibility into failed tasks and jobs

What is our primary use case?

We use it as a job scheduler with auditing capabilities.

We are using the latest version and host it on-premise.

How has it helped my organization?

It tells us when tasks fail in unexpected ways. Programmers test for common errors, such as when a job doesn't start. However, they may not be testing for when a DB deadlocks. This solution will monitor the return codes from all processes and alert us when something fails, whether or not a programmer has a test in the program to identify that problem. It has raised the visibility of these errors which we are working on to solve, making the code much more robust.

It provides more visibility to developers. It has given us better visibility into failed tasks and jobs, so we're able to start working on solutions before production starts calling. This has saved us money. We are now able to be a lot more proactive instead of reactive. We are able to solve jobs without people screaming and staring at us while we're trying to solve the problem.

What is most valuable?

We have created some workflows and find them very valuable. We are trying to move more of our tasks into workflows. Currently, we have about 850 tasks in it, and we are trying to move more of our tasks into workflows, but we have been slow on this. However, the workflows seem to be the most beneficial to us as a lot of our jobs are on different servers. E.g., on server A, we will run a job and its output will be consumed by server B. Then, on server B, we'll run a job on that. Its output will be consumed by server C. We don't really know how long these jobs will run on each machine. We just have a window. We expect server A to be done by nine o'clock, then we expect server B to be done by 11. After that, server C should be done by around one. In reality, that is overkill for most of the jobs. They don't take that long to run. So, if we can put them in a workflow, it just kicks off the next task on whatever server that it is scheduled on. In real-time, there is no dead time waiting on the files. The end result is users get much quicker visibility of things.

Stonebranch Universal Task is very capable. I like it. We have not really come up with a task that we can't do with it. While there are a few edge cases, 99 percent of our processing can be handled by this product with no problems.

What needs improvement?

While the resiliency is very good, it does have one point of failure that we are concerned about. We would like to run it in high availability in multiple clusters, but it has to read and write to one flat file. To us, that's a single point of failure that will prevent us from moving it to clustering like we would want to do. Therefore, we cannot run it on totally separate clusters and geographic separate databases in geographic dispersed areas. We would like to have one in Georgia and one located in Ohio with failover. Since the requirements are to write to one file, we just can't do that. That one file has to exist somewhere.

We would like the solution to work better with SSIS and SSRS. Right now, it just starts the job but does not give us any visibility into whether the job ran correctly or not. It tells us it started it, but it doesn't tell us how long it ran, any of the output, etc. We have lost that sort of visibility by going to Stonebranch.

For how long have I used the solution?

We started using this product in the summer of 2016. We had a fairly active migration until about mid-2018. So, it took us about two years to move everything into Stonebranch that we wanted to move.

What do I think about the stability of the solution?

The stability is very good and solid. I've had no problems with it.

The resiliency is good. We have had server outages and can tell whether to restart jobs that were running, pick up jobs that were skipped or just ignore them. It does a real good job of picking up after itself. In addition, if the server that UAC is running on goes down, it will do the same thing. It will try and catch up on the jobs that it needs to be in. It knows which ones to skip. It is a fairly robust solution.

What do I think about the scalability of the solution?

We have 17 servers that we're running in production which Stonebranch is communicating with. We have over 800 tasks running on these 17 different servers. I get the feeling it could scale to 10 times this amount with no problem. It does not seem to have any issues with the amount of jobs or data that it is handling.

We do plan on increasing it as new servers are added, but as of now, we're holding off on adding new servers. We have done as much as we need to right now. As new production servers are added, and if the Stonebranch capability is needed, we will add it. We do have more licenses, so it's not a license issue. 

I am the admin. We have six to eight users who are developers utilizing it and looking at tasks to ensure they have run or obtain data from failed jobs. Management does have the ability to look at it, but I don't believe that they are. Overall, we have approximately 10 users in the company.

How are customer service and technical support?

We have used the technical support fairly often. They seem to be very capable.

Which solution did I use previously and why did I switch?

This would be the first solution of its type that we have used.

How was the initial setup?

We had a consultant assist us onsite and the initial setup seemed fairly simple and straightforward to me.

The initial deployment was done within a week, then we started playing around with it. We were looking at individual jobs and determining if we needed them still and if they were in the correct form to be moved. We did a lot of code cleanup as part of our migration. To fully deploy and migrate our jobs, it did take us two years.

The implementation strategy was to move them over there, but we wanted them moved correctly. We didn't want to just slap them in and have bad code running on a different system. We wanted to ensure that the code was doing what we wanted it to, and that it was named properly. If it was sending an email, we wanted to send the correct email to the correct people. So, we started moving everybody into mail groups. It was just a lot of overhead that was not directly a problem for Stonebranch, but we wanted to get the jobs in Stonebranch in a good fashion. 

What about the implementation team?

We'd hired one of Stonebranch's recommended people. We probably could have done it on our own, but it's a lot nicer when somebody is holding your hand. 

The consultant was very knowledgeable guy and easy to get along with. He took a lot of time explaining things to me, even things I didn't understand, but made sense later on. It was a very positive experience.

They got us up and running in our production environment a whole lot faster than if we had done it by ourselves.

I do the deployment and maintenance of the solution. It's not a full-time job.

What was our ROI?

We are saving on eight to 10 hours a week of developer time, which adds up.

Which other solutions did I evaluate?

I am pretty sure that we did not evaluate other companies beside Stonebranch.

What other advice do I have?

It is a really great solution if you have your jobs and tasks well-defined and know what they are doing, It's a good solution for allowing you to shorten the windows between tasks. However, you need to get your back-end straight before you can start using Stonebranch to its full advantage. We had planned on doing all this stuff with workflows, but it turns out a lot of people don't really know why we run jobs at certain times when we do. So, we are having to rediscover that before we can take full advantage of Stonebranch.

The Universal Controller is very nice. It was sort of a foreign concept to me. It broke things down a little differently than I was used to or expecting. I come from a background where I used cron a lot. Therefore, I am used to the idea of scheduling jobs. However, with this product, everything was so interconnected with good error reporting. It was just a little new with a learning curve. It seems to be very capable, but it took a change of mindset.

As long as you're using standard Unix or Windows Task Scheduler, compatibility with other vendors is fine. It does not interoperate well with proprietary scheduling systems used by other groups. We have Crystal, which might have some problems interacting with them. There is some other proprietary stuff that we are really working with Stonebranch on trying to get a better interface with. However, as long as you're using standard out-of-the-box Unix or Windows Task Scheduler, this solution is rock solid.

**Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
Add a Comment