Ezriel Gross on the Importance of CICS
Reg Harbeck talks with Ezriel Gross about his history with CICS and its continued importance in the mainframe community.
By Reg Harbeck05/01/2020
Ezriel Gross: Well, to be honest, I graduated in the mid-80s and at that time there principally was only mainframes, so when you think about what you could have worked on back then, if you wanted to go into data processing or computer science, you ended up writing software on a mainframe. I started my career, believe it or not, as a CICS® system programmer for a bank at the time.
Reg: That is so cool. I started out as a CICS systems programmer in 1987 about half a year after I graduated from my degree, so I think we have a very interesting parallel path here. I worked for the Alberta Government in Alberta, Canada for my first couple of years before I moved to Calgary. Given that, I mean, where you were you located as you started out with your degree in your first job?
Ezriel: So, I actually graduated a year before you in '86, believe it or not. That's when I started, the summer of 1986, I worked for a company called Manufacturers Hanover Trust, which of course doesn't exist anymore, swallowed up by a bunch of other banks. I spent about two years learning CICS systems programming, did a lot of Assembler programming at the time.
Reg: I bet you wrote the autoinstall exit.
Ezriel: Yeah. At one point in time or another I've certainly written the autoinstall exit, that is for sure, but then I went to an insurance company where I did spend a lot more time. I spent about seven years at Met Life. Again, our group was a systems programming group over four datacenters so we kind of decided what would actually run in the individual datacenters. There weren't really that many sysprogs on staff in the datacenters, and so we made global decisions for the company, wrote some more software, decided what applications, or I should say systems vendor products, that we should run for monitoring and other things, and we were in charge of installation and maintenance and that kind of things, upgrading CICS, that kind of stuff.
Reg: Now when you were writing, I'm going to guess a lot of what you wrote was probably in Assembler but maybe some other languages as well?
Ezriel: Yeah, so actually what happened at Met Life was, besides doing systems programming, I got involved in some early IBM products at the time and some of them were application related, so I had to sit down and of course I knew COBOL.
Ezriel: Because I had learned a little bit in college and as part of my initial training program, I learned COBOL, so suddenly I found myself in this world of writing COBOL programs. I ended up doing more of that after leaving Met Life so to be honest with you, I was at Met Life for about seven years. Then I ended up at a company called Circle Education.
Reg: I took courses from them. That's so cool.
Ezriel: Yes, so Circle Education became Circle Computer Group and eventually Circle Software.
Reg: Their poster was one of my treasures, the CICS–
Ezriel: Oh, yeah.
Reg: Internals poster. That was one of my treasures.
Ezriel: We still have the internals poster, so I've written a few of those.
Ezriel: What happened was working for an education company, you meet a lot of customers, and they had a lot of consulting needs. A lot of the times, it was in the application area, so I ended up writing a lot of software actually in COBOL as well. It was mostly again COBOL CICS related software, some exits which of course were mostly in Assembler, but I've also written some vendor products that Circle used to market and sell at the time. I've been or I had been with Circle for well over 20 some odd years.
Ezriel: And then we partnered with a company called Fundi which wrote a bunch of IBM CICS tools. They're out of Perth Australia, so what happened–
Reg: That's a lovely city.
Ezriel: Oh yes, really nice, and what actually happened was is they didn't really have anyone here in the US that did CICS related stuff, so they said you know what? You're an education and consulting company. You have some software products. We'll acquire you and then you'll help promote some of our software products as well, and so I've done that for about ten years or so.
Ezriel: Just recently about a year and a half ago we were acquired by Rocket, so Rocket purchased Fundi and thereby Circle as well, so I continue to market and sell IBM CICS tools, CICS performance analyzer and CICS configuration manager for IBM through my affiliation with Rocket of course. But we also have a new tool that's just come out which is called C\Prof. It collects CICS trace but it doesn't run in the CICS region.
Ezriel: So, if you have trace running, we can extract it using cross memory services and provide you with that level of detail without interfering with the overhead of the CICS region itself.
Reg: Oh, nice.
Ezriel: So, I spend most of my time showing people how the tools work and I help customers with various different problems, sometimes do ad hoc consulting engagements and a little bit of training through conferences and presentations and such.
Reg: Now of course alongside all of that, you've obviously had a pretty substantial involvement with the mainframe community. Otherwise there's no way you would have been chose as an IBM Champion. How long have you been an IBM Champion?
Ezriel: Since the beginning of the program, I think.
Reg: Oh, my goodness.
Ezriel: So what, three or four years or something like that? I was also an IBM Gold Consultant for many years through my current affiliation through Rocket. We're an aligned customer and so therefore you know I'm not a Gold Consultant anymore, but I've been a Gold Consultant for well over 20 years, and so when they were building the Champion program and they were looking around for individuals in different areas, CICS being one of them because I've done so much in terms of training and consulting that it became a natural fit to become an IBM Champion. I've enjoyed the program very much, actually.
Reg: Uh-huh. Yeah, we've had an opportunity to talk to a number of people about the Champion program recently and we're going to do a lot more of it because there's just so much need to learn about from the people who have been in the program. Now that said, I'm going to guess again that the Champion program in your case was essentially a recognition of a whole lot of stuff that was already going on including in the community. I assume you're already involved in like SHARE and other user groups.
Reg: And conferences and such?
Ezriel: Yeah. I do many conferences a year. Besides for SHARE, I do the IBM Technical University conferences. Generally, we get a session or so at that conference. I've done some of the other Guide SHARE Europe, for example. I've done for about ten or 15 years. I usually do a CICS session or two and I've done some in Germany. I've done Guide SHARE Nordic at one point in time, so I've done a lot of presentations but you have to remember, I come from education, so I've been training CICS for 30 or so years. It's very easy for me to extract a piece of it out that looks interesting that I can present to a set of customers in the room including IBM'ers and hopefully they learn something from the session. When I present at a conference, I'm trying to do a little bit more than you'd see in your average presentation, something new about something. I'd rather do it more as a training session so that if you're learning about Liberty inside of CICS, I want to actually put together an example so that you can take some of my artifacts away and build your own Liberty JVM in CICS, run a sample, and feel like you've accomplished something. So to me, I try and put a little education behind it, because you see what it is today. It's very hard to get education in the mainframe arena, but that's simply because we're a little bit smaller than we used to be, and today's world is all about, you know, you could log onto the internet and find everything for free and YouTube and what not, so there's less of an emphasis on education. But in reality, there should be more of an emphasis.
Reg: Oh yeah, well, especially mentoring style, because we've got so many new people coming onboard.
Ezriel: Exactly, so there's a lot of mentoring that people get to do, but again to learn something new. You know you want to hear about it at a conference so we just finished that IBM z/OS® Connect EE session, right? So, we had a roomful of people who wanted to learn about it that never heard about it before. Some did, some didn't; they wanted to try and get something new out of that so we try and make these conference presentations more like a little bit of education, so that you can feel like when you walk away you can build something. You can solve a problem. You can do something with the information you've collected at the conference presentation, so I try and do that a lot.
Reg: Now that's really interesting. I have to say my journey is parallel enough to yours that I'm going to guess that one of the experiences that you've probably also had that I've had is really being fascinated by the history of some of the quirks of CICS that are cultural. The first one being the deliberate mispronunciation, because the word Six is already taken in English, and how it is pronounced in other languages and just the fact that you know just slightly over half of people in the US say C-I-C-S or as I like to say C-I-C-S (Southern drawl).
Reg: But that slightly over half of people in the US also say “kicks.” Then you go to Britain, it's “kicks” and then they pronounce according to their local rules in other countries. I have my theories about that, but rather than sharing them I'd like to hear your thoughts on that.
Ezriel: I've traveled the world, fortunately or unfortunately teaching CICS, and again in Europe, it's known as CICS, as you know. In America we tend to use the C-I-C-S and CICS interchangeably because Hursley comes over here so often and visits clients that people have heard the term CICS, so that's quite common. But you go to South America and it's Six. It's very interesting because when I haven't been there in a while and I go see a customer. They're talking about Six this, Six that.
Ezriel: I'm going Six, the number six? Is it something special in the culture? But if you go to Italy or Spain, they call it Chicks.
Reg: Right, I'd heard that. Yeah.
Ezriel: So, you're right. Depending on where you go, it has a different pronunciation and you have to get used to it because you don't realize they're talking about the same thing that you're talking about but yeah, I've seen at least those four different ways of pronouncing it.
Reg: Now one of the pleasures I have yet to have, and I really hope somebody listening to this will change that, is to actually visit the labs in Hursley and meet the people who keep CICS alive and well. Have you had a chance to do that?
Ezriel: Ohh. I go over there more than once a year, so there is an ISV conference. As a vendor, I get to hear about what's coming out in new releases a year or two before it comes out and they try and get our feedback, plus we have to make sure that our products will run day one with the new release of CICS, so they have an ISV conference every years. But we have a close relationship with Hursley, or at least I've always had one back at Circle, at Fundi and now at Rocket, so I actually go to Hursley several times a year.
Ezriel: Meet with different people, talk about some opportunities, try and help them where I can, so yeah. I try and support the platform very much.
Reg: Now of course as those of us who work with CICS are aware, all of the programs and all of the messages that CICS issues all begin with DFH and that's got a sort of a post derivation, as it were. I think the DFH didn't actually mean what people say it means now, but have you heard what they say it means?
Ezriel: Yeah, so let me just say, you know, and to clarify it, I've talked to many different people. DFH was just the next three letter acronym that was assigned at the time CICS was created back in the summer of 1969 but over the years, I've heard so many different ways to say DFH, right? Dollars for hardware, dumb, fat, and happy, destined for Hursley, so I've heard many different ways to remember DFH is the way I put it, but in reality, it doesn't really stand for anything. Sorry.
Reg: Well, my favorite will always be don't forget Hursley, but–
Ezriel: Yeah, that's a good one, too. Don't forget Hursley, but again, it really didn't stand for anything. It just happened to be at the time that it was DFH.
Reg: Now we as users of CICS and programmers all that, you know we kind of take it for granted, and the rest of the world has no idea it's there, and yet some people have asserted it's the most important software ever written. What are your thoughts on that?
Ezriel: So, I would say it comes as the second most successful software product that IBM has ever –
Ezriel: Written so the question is, can you guess which is the first?
Reg: I want to say z/OS®, but?
Ezriel: Would you believe something even simpler? COBOL the language.
Ezriel: It’s the most successful software product IBM has ever had, and I'm talking in terms of revenue and profit, right?
Ezriel: CICS is #2.
Ezriel: So, it is #2 which is not bad.
Reg: Especially because it's really hard to run CICS without COBOL.
Ezriel: Yeah well, exactly although you could write Assembler based –
Ezriel: CICS programs, but nobody does that anymore.
Ezriel: And you'd support C and C++ and Java® today, so they're keeping up with the times, but at the end of the day, CICS was and is very successful. If you went to the keynote here at SHARE, they talked about the fact that we run, I think it was 1.1 million transactions every second, so you can imagine the number of billions of CICS transactions a day. Every time you swipe a credit card, all the back-end processing, all the banks use it, the brokerage houses, the insurance companies. They may not like to talk about it, but at the end of the day, it's their core back-end systems. It's where all their data is. It's reliable. It's fast and it's secure. In today's world, you can imagine that's extremely important.
Reg: Oh yeah.
Ezriel: So, you know, they might move certain pieces off that are not necessarily good candidates to be on a mainframe. Everyone is trying to cut costs or do something different or they have a set of programming staff that know how to write programs in Java and use the new IDE tools, and so they'll move pieces of it off. But I tend to find, as this IBM z/OS Connect session that I just finished does, is we're always trying to go up to the mainframe, run an application and get some data back. So, at the end of the day, it is the repository, so to speak, of all of our data, and again could it be moved? Are there other options? Possibly, but it's going to take many, many years to rewrite the billions of lines of COBOL that run in production every day so I – no offense to anyone –
Reg: And why would they want to do that?
Ezriel: Yeah, well, but even if they do–
Ezriel: Let's just go down that road and say they want to rewrite it for whatever reason, right? You and I and our probably children and grandchildren are going to be long gone from this –
Ezriel: Earth before they get rid of all the–
Reg: Oh, yeah.
Ezriel: Software that's already written, and like you said, what do you get when you spend billions of dollars.
Ezriel: Moving a major application off the mainframe? You get the same application you had before and you’re just a couple of billion dollars poorer, right?
Reg: Yeah and if you're lucky, you have the same application. Chances are you had to cut a lot of corners.
Reg: Just because nobody else has what the mainframe has.
Ezriel: And you won't get the speed and you won't get the reliability.
Ezriel: And you know even with all the new stuff you hear about cloud, right? What's going to be in the cloud is going to be a mainframe, right? It's going to be running the same applications–
Ezriel: You had before.
Reg: Just somebody else's mainframe.
Ezriel: Right. So, you can't replace what you've built in 30 or 40 years overnight. It just isn't going to happen, and you could argue with all the new development plus moving this development, it would take more than 30 or 40 years just to get it off if you wanted to. So, again I'd like to say I'd be here in 40 years, but it would be a stretch, you know what I'm saying? Oh, I could be. It could happen.
Reg: Well, you know it's more and more common that mainframers just keep on working right into their 80s.
Ezriel: Oh, you should see it. I mean, I have a good friend that just turned 75 today and he is working every day, mainframe CICS person, so I see it all the time. Most mainframers tend to earn more and more money as time goes on. The more experience you have and the realization that it's hard to find that level of expertise just raises the amount of money people are willing to pay you, so it's a great place to be.
Reg: Yeah, yeah. I think that's going to be interesting to see. Now this has been really, really good. Any final thoughts you had before we sort of finish up?
Ezriel: Yeah, I mean again I'm enjoying myself in this industry. Like you, I'm fairly young right? We're the young ones, right?
Reg: Yeah. Oh yeah.
Ezriel: It's kind of strange.
Ezriel: And so, I think the next ten years are going to be the most exciting. Yeah, so that's what I look forward to.
Reg: Okay, great. Well, thank you so much Ezriel.
Ezriel: You're welcome.
Reg Harbeck is a mainframe enthusiast who has worked IT and mainframes for over three decades. He's the chief strategist at Mainframe Analytics ltd.
See more by Reg Harbeck
Sponsored ContentAchieve Compliance Without Impacting Productivity