What is our primary use case?
Over the years, I have worked with various firms to install and configure this system. The primary use case is automation, but the file transfer feature is also a valuable integration. Many people quickly understand the file transfer aspect because it is easy to set up. Therefore, automation is our top priority, followed by file transfer as a close second.
How has it helped my organization?
The solution has positively impacted my organization by allowing me to point to a single screen and say, 'That's the state of play in our batch operations.' Most sites do things in the daytime, so overnight, there'll be some batch processing running, and then they'll get file updates in the morning. If you can just look at the screen and see we're running 30,000 jobs overnight and this is where the problems are, you can just point to the three little red boxes that are what we need to fix. It's a very good way of being able to tie everything together and say, 'We know what's running, what's not running, and where the problems are.' Also, statistically, you can go back into history and say, 'This thing has been failing several times in the last three months, and we can see where our issues are and where we need to concentrate.'
What is most valuable?
The best features of Control-M include the fact that you can easily connect different platforms. For anybody coming to it, if you needed to script a solution that connected Windows and Unix and mainframe, that would be difficult. But with Control-M, you can just sit back and connect a COBOL program running on the mainframe, trigger something on the Windows platform, then do a file transfer on Linux. That's all basically just drag and drop. You can just indicate that these lines between these boxes perform those actions. The fact that you can tie different platforms together, including the cloud now, which many people are starting to use extensively, is very helpful.
Integrating Control-M with technologies for Data Ops and DevOps processes can be easy or difficult depending on the site. Many people have embedded methods for these things, so it can be difficult sometimes at various different sites to get it to integrate. It's not the tech or Control-M itself that's the issue; it tends to be the processes at different locations. It is very good at handling DevOps and the change management on the operations side.
What needs improvement?
Areas of Control-M that have room for improvement include the reporting feature. The reporting on Control-M hasn't changed much over the years, although it is in a different internal format. It used to be Crystal Reports, and now they've upgraded that. It would be better if that was really flexible where you could define your own reports. You can customize it a little bit, but when people come in with complex questions, you should be able to use that tool and access anything in the database.
Control-M has two internal databases that are core to the product. You can go in and do your own SQL queries against the database, but this reporting tool should really be able to do everything that you can do with SQL, and give you good information. Instead, you end up having to export to spreadsheets and then change and update them. It can be very labor-intensive to get this information out.
Other than the reporting, they've addressed most things over the years. Control-M is a tool that's been around for more than 30 years, so they have actually fixed most issues that you would encounter. They have a request for enhancement process that most users have sent requests to, but it doesn't move very quickly. The other challenge is they're supporting so many different platforms; BMC just wants it to be a trouble-free release. When users request new features, such as improved reporting, BMC's priority remains maintaining a clean-running system.
For how long have I used the solution?
I have been using the product for approximately 24 years.
What do I think about the stability of the solution?
Control-M is very stable. I would rate it a nine out of ten. It very rarely has anything that goes wrong. It has to be stable; if it wasn't stable, then we wouldn't be using it.
What do I think about the scalability of the solution?
Control-M is scalable for business requirements and needs. On the SaaS side, they've given certain limits which wouldn't allow for moving the entire batch currently to their cloud-based solution. But on-premises, it is completely scalable. Other customers are running a million jobs per day without issues. If you did have issues, you could add servers on the side that would handle the extra workload.
We have around 20 active users logging into the system daily. These users are responsible for managing the cache within the system. This leads to approximately 500 users who receive the messages. Additionally, there are millions of customer names involved. In summary, there are about 20 users logging into the system and around 500 business application users receiving the data.
How are customer service and support?
Their support is excellent. If you encounter an issue that affects your business, you can call them at three o'clock in the morning, and they will help you find a solution.
I would rate the technical support a solid nine out of ten.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
I have always used Control-M. I've worked at sites where they've installed Control-M as new and then migrated various things from CronTab. CronTab is just a Linux tool that comes with the Linux boxes to schedule, but it's the Linux native scheduling. It tends to be quite simple these days. BMC supplies scripts to do the conversion. Problems start to arise when people have created their own solutions on their own platforms, where 90% of the stuff is easily migrated, but 10% will be something specific to Broadcom or other schedulers, requiring script changes in Control-M. The same situation occurs when moving from Control-M, as people often leverage specific features that become problematic when trying to move to different solutions.
How was the initial setup?
It's on-premises. The organization I work for has IT operations in about 60 different countries around the world. We use Control-M in at least 15 or 16 of those locations, and each site can be quite different. For instance, in Paris, they only use it for the mainframe. Our main office is in Germany, where they are starting to use the test version, and they would like us to explore the SaaS version as well.
If I were to start looking into this, I think a hybrid approach would be the most logical way to organize things. This would allow us to have a clear view of both our on-premises and cloud environments, enabling a gradual migration to the cloud instead of making a major switch all at once. I've seen BMC presentations on this, and it looks promising. The advantage is that I wouldn't have to worry about where each workload is processed; everything can be monitored and managed from a single screen, which is very powerful.
Additionally, I've spoken with users who utilize cloud workloads, and their primary concern is cost rather than the specific platform. Control-M is effective at managing this. For example, if you're planning to deploy a workload on a specific server, the scripts can easily handle the transition to Amazon's cloud the following week. If a provider has a special offer in a month, like Google, it's straightforward to switch our workload to run there instead. With Control-M, this process is transparent and uncomplicated.
The maintenance is complicated. There is patching to consider on both the Control-M service side and for the agents and modules. When it comes to file transfers, you need a specific module from the Managed File Transfer (MFT) service. Typically, every few weeks, we need to install a new set of updates, whether due to a security issue or the introduction of a new feature. If you go to the Application Center (AC) and use that version with your account, you don’t have to worry about caching, as they manage that on the systems. You could use it in a controlled manner, perhaps on a Sunday morning. However, using a basic version might present some challenges for us.
Usually, the Linux team handles the patching, but it's also necessary to conduct tests. Sometimes, the latest version can alter certain behaviors that might not be acceptable in a production environment, so testing is crucial. It does require a separate environment dedicated to that. Overall, managing patches is a significant task. One key selling point of the SaaS solution is that you won’t have to worry about it in the future; BMC takes care of that for you. However, some of my colleagues are concerned that implementing a SaaS version could mean losing a significant portion of their work to automation, particularly the patching responsibilities. It’s definitely a complex situation.
What was our ROI?
It's hard for me to determine, but at the current location, we did conduct a study. We use Control-M to run a data warehouse, and an external company evaluated our setup. They concluded that the way we operate with our existing tools and at our current scale is actually more efficient and cost-effective than purchasing a packaged data warehouse product, like Informatica.
At my previous employer, we performed a similar study, and the results showed that the product paid for itself within the first year. Moving forward, it continued to be very cost-effective. The benefits tend to be hidden—people often overlook them because these processes run in the background, handling file transfers and other tasks. If I had to hire a developer to script these operations, those costs would be substantial, along with ongoing maintenance.
With Control-M, you define the processes once and then let them run. The operations team can monitor it, which means it integrates seamlessly with existing systems. In terms of return on investment, it pays for itself quite quickly. However, because these advantages are often invisible, people don't fully appreciate the costs and savings involved. You really need to consider the expenses of hiring a developer or operations staff specifically for these tasks. If you use Control-M, those costs effectively disappear, and people tend to just assume that everything is functioning as it should. It's definitely worth noting the return on investment. I'm sure BMC would be happy to provide those figures, demonstrating that it quickly offsets its initial costs, even if there can be significant upfront expenses at times.
What's my experience with pricing, setup cost, and licensing?
Pricing for Control-M depends on the licensing model, with different options such as the per-agent model or the per-job model. BMC is phasing out the per-agent model in favor of per-job licensing, which can be expensive. The Control-M Helix SaaS version will only be available with per-job licensing. Currently, some implementations use a flat rate based on the number of agents. The shift to per-job licensing could have significant cost implications.
Control-M tends to be more expensive compared to other solutions, but users get great value from it. Organizations have invested considerable effort in developing for Control-M, making migration away from it less desirable. Competitors may offer fewer features at a reduced price, which could be a deciding factor for some companies.
Which other solutions did I evaluate?
Years ago, I was involved in a lot of Proof of Concepts (POC). Back then, IBM's Tivoli and Control-M were quite similar and difficult to differentiate. However, over the years, it’s become clear that IBM has not developed Tivoli to the level that Control-M has achieved. In my opinion, Control-M is now the clear market leader.
These days, you never know what new solutions might emerge that could disrupt the market. Nevertheless, Control-M has a robust footprint—it can integrate with many legacy systems, making it hard for competitors to come close to its capabilities.
While there may be newer solutions out there, I’m not fully aware of all the latest developments. Control-M does have advanced workload balancing in the cloud, but I still believe it remains the top choice for me. I may be biased since I've always worked with Control-M, but when you compare it to alternative tools, you often question whether they can accomplish specific tasks or integrate with various third-party products. You soon discover that many alternatives simply don’t offer the same level of functionality or integration, and even if they plan to provide those features in the future, they haven't yet delivered them.
Overall, I am quite confident in saying that Control-M is the number one product in the market. Of course, if you visit a potential customer site and they have unique requirements—especially if they are primarily Windows-based—they might find another product that better meets their needs than Control-M. However, in general terms and across most scenarios, I firmly believe that Control-M is the market leader in 90% of cases.
What other advice do I have?
Our relationship with BMC has been transformative. However, I get the feeling that we are moving towards a more transactional relationship. Previously, they were very concerned about us using the products, growing with them, and deriving value from them. This was a refreshing attitude from a software company. That said, they are restructuring, and I get the impression that in the future, they may want to accelerate revenue generation. So, the nature of our relationship could change. At present, though, it remains transformational. They have been very helpful, with excellent support, but who knows what the future holds?
I absolutely would recommend BMC Control-M to other users. My only reservation would be with the new licensing structure. If you are going to be paying per job and your batch process was very granular without the ability to restructure, then cost might be an issue. Also, it is the serious high-end scheduling solution, so smaller enterprises might want to consider whether they need all the features that come with Control-M. For small or medium-sized enterprises, different solutions might be more appropriate.
Overall, I would rate Control-M a nine out of ten.
Which deployment model are you using for this solution?
On-premises