What is our primary use case?
A lot of the things we've done are just based on our needs, not so much because the product allows you to do it. Basically, I can do everything in Control-M. I mean, we've got plugins for Oracle, SQL, and Informatica, and I can go on and on and on. However, we don't use any of them as our developers prefer not to. A lot of what they do is they do the necessary connections through the batch files themselves.
It's used for our daily batch. It handles all the batch processes and a lot of our maintenance processes. I would say most of it is file movement of some sort. A lot of it is daily processing, to get it in. Our data warehouse runs through Control-M. The big impetus behind it, when we purchased it, was due to the fact that the auditors wanted a more robust system and something that they could audit. Control-M gives you everything you need for that.
How has it helped my organization?
It allows us to automate a lot of the jobs that used to run manually. Everything is automated. We can automate a lot of different processes using Control-M. You can know where it's at, and you can follow it, follow the job flow, from one job to the next, and whatnot, very easily.
We used to run a lot of stuff in AT scheduler and Cron which really didn't meet the needs, especially for the auditors. We've taken that, and we've made the system where you know immediately if you got a problem with a job string. Our operations department will page it out overnight if we have a problem, and we take care of it. It's like any other system. If it allows you to do what you need to get done, it's the same every day, you know that you're going to get the same process. It drives the process.
Like most schedulers, you can bring jobs in many different ways. There are different ways to execute things. One of the things we had was when we were taken over. They were using a combination of the CA scheduler that they had, and they were also using SQL scheduler to do a lot of it. Prior to us converting our data warehouse system to Control-M, they were using the Informatica scheduler. None of this met any of the auditors. The auditors didn't like it as everything was spread out on different systems. They couldn't keep track of jobs. Everything's consolidated now. Everything's running off Control-M. You can follow everything through the entire process. We kick off all SQL jobs using Control-M. They were using SQL to launch just batch files, which had nothing to do with SQL - they were just scheduling it through SQL.
What is most valuable?
The capabilities of auditing have been great.
The ease of use is one of its great aspects. It's very easy to use and very easy to pick up.
It's got an excellent graphical interface. I haven't seen that in anything else that I've looked at, however, that said, I haven't looked at many lately.
I know that in 20 years, I have had probably two problems where I've had to call the company to get immediate assistance from them, where we had a system down or something. Its performance is very reliable.
It integrates with other applications. You can use PowerShell, you can use Perl, you can use whatever. It doesn't really care. It's just running a process.
The product scales quite well.
Technical support is very helpful and available 24/7.
The stability is excellent.
What needs improvement?
I will say that at one time we used to run on Solaris and not Windows, however, we were taken over by a company that decided that everything had to be on Windows. We put this in when we were the previous company, and then we were more or less given to the current bank by the FDAC, during the 2009 banking crisis. At that point, they wanted us to implement their solution, which was rudimentary at best. It was a CA product that did not meet the needs. I could not convert what we had in Control-M, to run in that system at that time.
While they have a very good reporting facility, the reports that I'm asked to produce, a lot of times aren't necessarily what we need. They need to be better customized. I haven't been able to produce the right reports through their reportings facility. I was a Perl programmer and a C programmer at one time. Perl just worked right in there. A lot of our reports were written in Perl, which right now they don't like at all as Perl's not ideal for our company.
I can't get to the database tables I want to get to. The database tables they allow me to get to aren't the ones I'm looking for, as, usually, I'm going right into the database, into the raw database, and pulling things out for the reporting I need. I can't do that through their reporting facility, Crystal Reports.
For how long have I used the solution?
We've been using the solution for two decades or so. It's been about 20 years. We've used it for a long time. We started using it around 2000 or 2001.
What do I think about the stability of the solution?
We've had issues only twice in 20 years. It is very stable. I will say that they have improved it. Originally, when we put in a Windows version of it, we had problems with the database that they were using at the time, which was a Postgres database. Then, at one point, we decided to go to Solaris and run it on Solaris. We had it on Solaris for six years. In six years, I don't think we ever rebooted the server. It ran for six years without any hiccups, any problems. The Solaris system was rock solid.
Now, the problems we run into, if we run into any problems, are Windows-based problems. Those Windows-based problems are, for example, if you don't reboot a server once a month, which, thank God we do, you can have issues. We reboot as we have to patch monthly now and we have to reboot it every month. However, we would see if we went two, three months on Windows, that we would start seeing some problems. Rebooting it took care of it.
That said, that's a Windows problem, not so much a Control-M issue, as we see problems on Windows servers that run for two or three months in any application.
What do I think about the scalability of the solution?
Right now, we are running on their small database model. We, at one time, had about 2,500 jobs, and we were on a medium model then. Now, we're down to about 800 jobs a day. It's just a matter of the requirements we have. In terms of scalability, it scales up very nicely. It works very well. You can have multiple servers if you need multiple servers. Currently, we have one Control-M server and one EM server. We used to have two Control-M servers and one EM - EM being the enterprise manager, which is really what's running the system. The Control-M servers basically take care of the current runs, what's currently running on a system. Adding more jobs and adding more resources to it is not a problem.
It does high availability. We don't use the high availability due to the fact that we have another solution. We run everything in a virtual environment, and take regular snapshots if the system goes down. Should that happen, the snapshots are replicated from our production site to our DR site. We bring up the latest snapshot in the DR site if we lose the production site. It's up and running within minutes, literally. It's just a matter of going in and saying, "Bring these servers up." And they come up.
Currently, we've got three schedulers using the solution. They have more or less God rights, although they can't change user permissions. Those three schedulers can do anything with the jobs - delete, add, create, whatever. We have about 10 operators that have access to it as well. The operators have a somewhat reduced role from the schedulers. They can do a lot of it. They can bring in jobs, they can rerun jobs, they can kill jobs, however, there's a lot that they can't do. Then we have probably about 60 users that are developers, and they're basically read-only. They can see the jobs, they can see what happens. A lot of it has to do with corporate decisions on control. They didn't want the developers to be able to define jobs and items of that nature. They wanted the developers to define the job through a worksheet, and then the schedulers would actually implement the job. That's just a matter of policy, basically. They monitor their jobs that way. I'm trying to allow them to be able to at least bring in their jobs, for test - not for production - so that they can make it policy change here. If they could do that, it would greatly enhance their ability to get testing done. The downside to that is that you might have a developer that just keeps running the job over and over, and over, and over again, which I've seen happen too. Personally, I can do everything in test. I can't do anything in production at all, except view jobs. I have read-only on everything in production, except for the configuration part of it, to which I have full rights. I used to almost be a fourth scheduler at one time. At this point, there's no need. The limits of my job have been redefined several times.
Overall, the usage of the product in the company is very extensive. There's not a part of our daily businesses that's not reliant upon Control-M. If Control-M was done, the company would be at a standstill, literally.
That said, likely we won't increase usage. The company we just merged with, another organization and it's debatable as to how these things go. They have about 5,500 jobs. We used to have a lot of jobs like that, however, the business drives what we do.
How are customer service and technical support?
The technical support is probably the best I've ever worked with.
If I need support help from them, if we are down, they get back to me, if not immediately, within an hour. 24/7. And usually, we're up within an hour, after the first contact. They help greatly with planning for upgrades. I need to contact them here in the near future. They have a group called the AMIGO group, that does nothing but migrations and upgrades. I need to get with them to go over my plans for transitioning from the old servers to new servers. They will verify that what I'm doing is the right way to do it. If it's not, they will tell me how to do it, which is an excellent resource.
They have a very large knowledge base. It's integrated with everything I've ever had to have it integrate with. Their support's been very good.
When I call BMC, I get an immediate response. I've had products that I've supported, that I've called companies and been on hold overnight. I've literally gone home for the night and left my phone on my desk, off the hook, on hold, and come in the next morning, and I'm still on hold, listening to the hold message due to the fact that the support hasn't answered yet.
Which solution did I use previously and why did I switch?
We have recently merged with a company that uses Tidal, and of course, they want to hang on to theirs. We use Control-M. I've actually used several other scheduling products in the past, however, we've been on Control-M now for over 20 years.
How was the initial setup?
I'm actually in the process of doing an implementation right now. I'm replacing our current production system. We're replacing EOS, actually, therefore, I'm doing a straight install of everything on the new servers. It is very straightforward. The install is not really difficult. It's fairly simple if you understand how databases work and whatnot. There's really no problem doing it.
In my case, I can bring up a Control-M server within hours. I only say that as I've done that, as we were not DR prepared back during Hurricane Sandy. I had to bring up a production version of it in Cleveland, in our DR site in Cleveland. Within 24 hours, we were up and running. Therefore, if you need it done fast, it can be done. It's just a matter of, are you willing to put in what you need to put in to do it.
It's a fairly easy install, really. I personally have never had any training on Control-M. Other people in my organization have had training. That said, I'm the one that put it in and I'm the one that read the manual. That's where I got all my information from, was from reading manuals and whatnot, and directly working with it.
What's my experience with pricing, setup cost, and licensing?
I can't speak to what our support costs are. That's out of my realm at this point. At one point, I had an idea, however, I couldn't even tell you what that is anymore. I know that our licensing is based on jobs. We buy licenses based on the number of jobs. Currently, we have about 2,500 licenses. We used to run more jobs than we do right now. We did not get rid of those licenses.
It's basically $100 a job, give or take.
They also don't charge us for items such as the plugins for MFTP, which we don't use, although we could. They wouldn't charge us for Oracle, SQL, or Informatica. It's a reporting product.
There's no licensing for the server, there's no licensing for the EM server. All that stuff comes as part of the product. It's all-inclusive.
From what I've seen and heard from the other company about Tidal, that's where they're making their money from - the plugins. Whereas Control-M doesn't charge us. The plugins are basically free for us. I'm sure there is a charge for support every year. I have no idea what that is. I don't get down into that level.
I just tell them, "Yes, we need this" and then the purchasing staff takes care of the actual details.
Which other solutions did I evaluate?
At the time we were looking for a product, I looked at five or six different scheduling packages. By far, at that time, Control-M was leaps and bounds above all the rest of them.
What other advice do I have?
We're customers and end-users.
We're using the latest version of the solution.
By far, BMC, from what I have seen, is the industry leader and they are the Cadillac of scheduling. I've worked with a lot of different scheduling systems over the years. When I first got into IT, years and years and years ago, as a JCL programmer, basically you had access to the scheduling system and you took care of the jobs. When jobs failed, you would do the restarts on them, do whatever fix needed to be done, and get them restarted, and get them to rerun. That was on a mainframe.
I've used Cron, and I've worked with a number of different schedulers. In the Windows world, other than AT scheduler and Control-M, that's about all I've ever used. I did review five different products back when we put this in.
Having worked with so many products, and with this one for so long, I can advise that new uses should follow the installation instructions and notes. They're very simple, very straightforward. I would advise others to not get scared off by the price as, initially, the pricing seems rather steep, compared to some of the others. However, they all have their pricing quirks, and they're all making money in one way or another. The way they make their money is based on the way they license it. The per-job style actually works out very well.
I'd rate the product at a perfect ten out of ten. It has been one of the most stable products that I have supported, and I have supported a lot of different products. I've had fewer problems with it than I have with just about anything else I've supported.
Which deployment model are you using for this solution?