Hemanth Rama: I am a Mainframer

THE OPEN MAINFRAME PROJECT

Jeff Frey: Welcome to another edition of the “I Am a Mainframer” conversation series sponsored by the Open Mainframe Project. I’m Jeff Frey, and I’m a retired IBM fellow and previous CTO of IBM’s mainframe platform, and also a very long-time mainframe enthusiast. It’s my pleasure today to host another of this series in the mainframe conversation series. As a Linux Foundation collaborative project, the Open Mainframe Project is intended to help a mainframe-focused, open-source technical community. It’s also intended to serve as a focal point for the development and deployment of enterprise Linux in a mainframe computing environment. The goal of the project is to excite the Linux community around the use of the mainframe and to foster collaboration across the mainframe community to develop and exploit shared Linux tool sets, resources, and services in the mainframe environment.

In addition, some of our broadcasts show specifically that the projects seeks to involve the participation of academic institutions in order to help and assist, create educational programs aimed at developing mainframe Linux engineers and the developers of tomorrow.

So for today’s conversation, we have the pleasure of speaking with Hemanth Rama. Mr. Rama is a senior software engineer at BMC Software and a mainframe expert specializing in performance, System z software costs, and the z/OS Workload Manager. He regularly writes for many of the many popular IT websites and also on his personal blog. He passionately speaks about mainframe technologies at various conferences, symposiums, and user groups, and he’s contributed to various BMC performance monitoring-related products and is currently working on a product that optimizes System z MLC costs. Hemanth, welcome to the broadcast. It’s a great pleasure to have you on today.

Hemanth Rama: Thank you, thank you Jeff. Thanks for having me today.

Jeff Frey: You’re most welcome. This should be a great discussion. To get us started, can you tell us a little bit about you. Give us some background on your career, how you became involved with mainframe systems and maybe a little bit more of what you do at BMC?

Hemanth Rama: Sure, sure, sure. And once again, thanks for having me today, Jeff. It seems as if it’s a great show to share your story with fellow Mainframers, and thanks for running this great show. My name is Hemanth Rama, and I work for BMC Software in Silicon Valley, California. Regarding my career background, it all started out at Northern Illinois University where I started to pursue my Master’s in Computer Science degree from Northern Illinois University. This back in 2003, 2004. Back then, NIU had excellent mainframe curriculum and they also had some brilliant professors like Robert Rannie and Michael Stack. Some of them probably you know, Jeff. They’re very frequent at SHARE. And in fact, I met Michael Stack at last  SHARE in Rhode Island, Providence. So I can’t really underestimate the lasting impression I had from this brilliant professors who taught me mainframes at NIU.

And also, as a part of curriculum what we did was we developed a  Student Operating System short for SOS where we used similar tests to demonstrate the various MVS internal architecture constructs. So that had a big impression on me and I really, really enjoyed developing that operating system, Student Operating System. So I hooked up to the mainframe and to be  honest, when I almost graduated with 4.0 GPA, so that tells me that I’m pretty much into … How much I liked on the mainframes. So, and then back then NIU had these mainframe courses as a  required for you to graduate. They’re not optional, they’re required. But I’ve heard that it’s not the same story now, but back then it was required. So that is one good thing that going back to NIU that they made these courses required for you to graduate.

It doesn’t mean that they do not have the other curriculum. They do have the other main, other modern technologies, which I also took, which is Java, .NET and database servicing but it was mainframe that I felt more connected with. And from there on, once I graduated, back in 2005, I got an interview opportunity with BMC Software. So, I flew into Silicon Valley and after a couple of rounds of interviews, they offered me a job. And I liked  what they were doing back then and the team was great so I started working for BMC. At BMC, what I do is that I research and develop products. I primarily used to work on  MainView for z/OS and CMF monitoring and SYSPROG performance monitoring products. These are the products used by mainframe SYSPROGs worldwide to do their daily task. And then more recently I am working on a patented product called Intelligent Capping, short for iCap, which optimizes MLC costs and this is where I got two patents this year, and one pending application. To go in deeper, what I actually do is that I write application in Assembler, C, and some web-based technologies.

Jeff Frey: I see. So, your work up to this point has been mostly in the context of the traditional side of the mainframe, the z/OS side of things. Have you been exposed at all or, have you worked at all on the Linux side of the platform?

Hemanth Rama: Not exactly on the Linux, Jeffrey, but I did work on the USS Unix side.

Jeff Frey: I see.

Hemanth Rama: Yeah, inside of the mainframe you do have the USS so I got an opportunity to work on that as well, but not directly on the Linux, yes.

Jeff Frey: Yeah, I see. I see. I read that you have some experience with Workload Manager. That we, for a long time, have been pretty proud of the capabilities of WLM in a z/OS environment and I think it probably was the first and maybe even today is the best tool in the industry for doing goal-oriented performance management reporting. How much experience have you had with WLM?

Hemanth Rama: Yes, yes. I think what you just said, Jeff, that is perfect. I mean, the WLM is the most important component of z/OS and in fact I gave two sessions or maybe four sessions in this  SHARE from introduction to the advanced level of the WLM. So, I got into the Workload Manager when I was working on this performance products. We often deal with the Workload Manager with the policies  Service classes and everything. So, we extract data, we provide data, display what’s going on with the workloads. When we moved  to work on iCap which is the most recent product I working on we went even further in using the WLM to get the rolling four average and helping to feeding back the data for the  Defined Capacity and everything. So, the best thing I’ve worked with Workload Manager is how it manages tons of, tons of transactions … And a simple mainframe, that’s simply amazing. The architecture, everything. It’s just awesome.

Jeff Frey: Yeah, yeah it’s of course as you probably know, it’s been developed and continues to evolve over many, many years. At first, I think clients were a little bit reluctant to use WLM in goal mode because people were so used to setting the knobs and dials on performance in a very manual way. And, didn’t really believe that the system could monitor itself and make adjustments according to a specified set of goals and be good at it. But I think over time, it’s proven itself to be just the crown jewel of that platform. I agree … I wish more people had an appreciation for WLM and its capabilities. It’s really a marvel of modern engineering.

Hemanth Rama: Indeed, indeed. I agree.

Jeff Frey: Let me ask you given that I think we share an enthusiasm for the qualities of service, the mainframe, and obviously in the area performance and economy of scale and people are fairly well-versed in and recognize the mainframe as an available platform, and a secure platform. And it has those attributes. What do you think, on the flip side of this, what do you see as the biggest challenge or obstacles for the mainframe going forward?

Hemanth Rama: Yeah, that’s a good question, Jeff. I think what I observe I mean based on work I had been in the industry is that I see that the baby boomers and Generation X, those are the excellent people who kept mainframe running and inventing and evolving. These people are retiring and I don’t see lot of millennial’s stepping up into their shoes. So I feel that there is a significant shortage of skills … Skills gap currently in the mainframe. But thanks to the several initiatives taken by IBM, and the other vendors like my company BMC, programs like Masters  the Mainframe or the Academy Initiative we are cooperating with various universities, and zNext Gen community within the SHARE conference. These are some of these initiatives and there are other things like BMC where they are encouraging also the younger people to hire and work on. So, these are the few things I am seeing, initiatives taken by various industries to just fill this skill gap. But it’s a long way to go, I mean it’s a good start but it’s still a long way to go.

Jeff Frey: It’s an interesting discussion that I get into with people on skills, because certainly there are many systems in the industry that have their own personalities. And they express their capabilities and especially in terms of how they’re managed, how they’re configured, managed and the operational controls of the system, the operating system, and the middleware can be very product-specific. But I don’t think that’s any worse or any more of a problem for the mainframe than it is for other systems in the industry. And I think the thing about skills is that people can learn how to do things, the mechanics of configuring a system, the operational controls of the system, how to manage a system, how to secure a system, how to make a system available. The mechanics of doing that are something that are very learn-able, right? Even though they might be different from platform-to-platform in some ways.

I think the real emphasis on skills and training should be not so much on the particular language or semantics that are used in managing a system, but the concepts of availability, the concepts of disaster recovery and security in this day and age. And this is where the mainframe excels. I think we would do ourselves a service if we focused the education and our efforts to educate people on enterprise-class computing concepts, right? And how to make IT shops available and low-cost and efficient and secure and available and disaster recoverable. And then, right, lead them to conclude naturally that the mainframe is the best platform to serve those needs. And learning the syntax of how to do that, right, and the policy and procedure and best practices on how to do that is … People will get excited about that then, right? I think some of the education we do is aimed a little bit too at these lower levels of things, and we focus on syntax rather than the conceptual notions of enterprise. Yeah.

Hemanth Rama: Yes, yes. Like you said Jeff, the platform is strong. I mean the foundation is pretty strong. It’s just that we need to spread it so that we can  attract like a certain higher level, the conceptual level, architecture level where we can attract more and more and what I’m seeing is that once they get a taste of it, they’re really, really and do so many people when I talking to younger Mainframers.

Jeff Frey: Yeah, yeah so that’s cool. Let’s jump into this next … It’s kind of a natural lead-in from our last discussion. Let me ask you what advice you might give to other IT professionals who could benefit from the use of the mainframe in their shops.

Hemanth Rama: Yeah, I think if you really wanna see the benefit of the mainframe technology, right, the best way to start out is attending one of these conferences like SHARE or CMG where you get to see what the current mainframes are. I see these conferences as a continuous education. You have the foundation, but these are the conferences where we learn a lot of the latest and greatest innovations happening across the mainframe so that you could know … You could benefit from one of these. Every time any new hardware or software comes in, it’s guaranteed that it’s making a biggest improvement than what it used to or previously. So, I strongly encourage the fellow Mainframers as you keep attending these conferences so that you can keep up with what go on. The other thing is that not just this once or twice a year conferences thing, but you could also do year-round activities like we do as part of zNext Gen community from SHARE where we have a monthly webinars. So, we introduce the new concepts to the other mainframes and it’s free to join anybody. So we introduce this concept of the mainframes year-round, monthly basis. Some of these that I would encourage to attend.

And the other thing like blogs, articles, and interviews like platforms like yours. There are several other platforms like this. Nation Z, Plat … Mainframes. So these are the platforms that I encourage people to go on, check it out because there might be somebody writing there on skills or their own knowledge sharing, everything. So this are some of the other venues I would encourage I would encourage my fellow Mainframers to explore.

Most recently, what we are talking about Jeff, is that if you look at the other technologies you have some just common platform where they go and ask questions and get help for. So we do have certain platforms where we ask questions and get answers for mainframe but it’s not that we don’t have … I think a single platform where we say, “Okay, this is a place where I can go and talk and ask a question and get a feedback on” … That’s still not there yet. So one of the things we as zNext Gen doing is that we are trying to come up with at least one platform where anybody can go like from junior to more experienced. So they all come on one umbrella so that people can exchange their questions and answers and views and etc. So one of those things that we are working on. So these are these few things that could help benefit those using the mainframe.

Jeff Frey: Yeah, that’s great. I also wanna say that I really appreciate people like yourself who have so much enthusiasm and have so much understanding and appreciation for the value of the mainframe. It’s broadcasts like these and other places that people like you and I can offer testimony for this great platform. I run into all kinds of people where they either have an older or a historic and almost antiquated view of the mainframe because they haven’t necessarily kept up with the advances and the evolution of the platform and so I think a lot of people are actually surprised when they learn about how much modern technology and state-of-the-art technology hardware and software are in this platform. And maybe they assume that because it’s got its architectural roots back in 1964 that it hasn’t advanced, but we gotta continue to spread the word and make sure people understand just how great this thing is.

Hemanth Rama: Yes. That’s critical.

Jeff Frey: All right, hey listen it’s been a pleasure talking with you. Maybe we can find a way to meet face-to-face at some point and I surely do appreciate you expressing your point of view here and helping us out with these conversation series. It is much appreciated, Hemanth. Thank you very much.

Hemanth Rama: Thanks, thanks a lot, Jeff. And this is indeed great to share with fellow Mainframers. I appreciate.

Jeff Frey: All right, so that does it for us and we’ll see ya next time on the mainframe series. Bye now.

Hemanth Rama: Sure. Thank you, thank you.