IDUG Db2 Table Talk
IDUG Db2 Table Talk
Things I Wish I Knew before Interviewing Michael Cotignola
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
Famous for his globally recognised presentation, ‘Things About Db2 I wish I'd Remember When’, BMC Software’s Principal Solution Engineer, Michael Cotignola is the latest victim of Marcus and Julia’s latest podcast.
Learn more about IDUG, the International Db2 User Group, at www.idug.org.
Hi everyone, thanks for joining us for this series of iDug Podcast, where we speak to some great people in the DB2 community. I'm Marcus Gavage.
SPEAKER_01And I'm Julia Carter, and we represent the iDug Content Committee. We hope you enjoy today's edition.
SPEAKER_04Welcome everybody. Thanks for tuning in. Today we have Michael Cotignola, IBM Champion, iDeck Content Planner Committee member, and all round good egg. He works with BMC Software as a software consultant, and he has the honor of being a good friend. Welcome, Michael.
SPEAKER_02Oh, thank you. Thank you. Thank you for having me. Glad to be here.
SPEAKER_01Michael Kaylee, you've been working with DBT for a few years, let's say.
SPEAKER_02Actually, when when I start when I sat down and thought about this, about what what we talk about in on this little talk of ours. Yeah, a few years is about right. It's it's staggering when I think about it.
SPEAKER_01I was doing the math and I was I was trying to work out like I don't think you were born yet. But yeah, the the math isn't mathing as the kids are saying these days. But you've been working with DBT since version 1.1.
SPEAKER_04Correct. Correct.
SPEAKER_01Right. And and what were you doing back then? I mean, how did how did you become part of of of version 1.1 of DBT?
SPEAKER_02So uh I was working for the local phone company at the time, uh, and we were redesigning uh a major application, uh, and the manager at the time thought, huh, why don't we take a look at this DB2 stuff? So basically, we were kind of like a beta program, um, and we did everything IBM told us not to do. Because at that time, no, no, no, no, no. DB2 is not for production, it's only for ad hoc reporting, definitely not a batch system or whatever. So we ended up doing everything they told us not to do, and but the project was successful. Successful in the sense that it was on budget and on time, which is you know always a good thing. But from uh a technology perspective, it was very, very successful. So basically, I was assigned to the project. None of none of us knew anything about DB2, so it was quite the learning experience. Uh I I think I probably learned more about DB2 back then than I have in all the years cumulative since. Because you you always learn when you're doing hands-on stuff, right? When you're just reading. Uh yeah. But when you get your fingers burned a couple of times, that's when stuff sinks in. So it was it was the general ledger system for for the phone company, which was quite uh ambitious for the project manager at the time to undertake, because you know, the phone company needs to close the books at the end of the month, and if the application isn't working, they don't get their money. So but it was good. It was good. It was quite uh, like I said, quite the learning experience. And it was quite advanced at the time in the sense of believe it or not, we had the end users do their own reports and they used QMF to do it. So business analysts wrote their own reports using QMF. So that was also something quite we thought innovative at the time.
SPEAKER_01Yeah, no, that's pretty cool.
SPEAKER_02Yeah, yeah.
SPEAKER_01And so were you a developer of I was a developer, correct.
SPEAKER_02Yeah. I knew nothing about databases. Uh so I was a developer. Um and one of the things I learned about, and it's been a running joke, and I think a lot of people use use it in a lot of presentations when they're talking about DB2, uh, in terms of uh you know the application programmer and the the fear of committing, right? So I learned the hard way that commits are something you should be doing. Because the first program I did was a badge program, right? Uh and it was an update program. So I started doing my unit testing. 10 transactions work like a charm. Okay, let's try some high volume. Did some high volume testing, and all of a sudden uh the program would abend. That as far as I knew, right? Okay, well, what happened here? I don't know. In those days, just resubmit, right? Anyway, make a long st make a long story short. It took me a while to figure out that db2 was crashing at the time. And it was crashing because the RLM didn't have enough storage for all the locks I was accumulating. Because I wasn't doing any committing. And this was way before you know lock escalation and all this other good stuff that's in DB2. So it took me quite a while to figure out, hmm, I need to commit. So that way, you know, I release my locks and the RLM doesn't come crashing down. Yeah.
SPEAKER_04So since since then, DB2 has come a bit of a bit of a ways. Uh oh yeah, yeah, it has.
SPEAKER_02But yeah, like I look at the stuff that they're doing now, it's like it's so uh out there for me. I mean, I mean I'm still going back to wow, packages, that's such a cool thing. And you know, if you think about it, right, it's like so this was way back when. I I forgot what version of DB2. Yeah, packages came out in 2.3. That was the first thing I saw. Oh my god, okay, yeah. So so that was quite a ways ago, right? And this was when, you know, like um waterfall development and modular programming were still uh vogue, right? And it's like modular programming, and then you look at well, why do I have to bind this monolithic plan with all the DBRMs in it and it takes forever? And then IBM comes out with packages, is huh? What a brilliant idea, right? Yeah, so actually similarly with data sharing, right? Like so, data sharing IBM touts it now for high availability, which is probably true. But as far as I can tell, or my perspective on that, IBM came out with data sharing because customers back at the in the day were throwing so much workload at a DB2 subsystem that the hardware or the the the machine couldn't cope with it. So they needed to be able to run DB2 on multiple machines, and the only way to do that is data sharing, right? So great technology, great invention, great idea. Fast forward to today, and people a lot of customers I'm working with now are basically destroying data sharing members because the machines have gotten so fast, the capacity's gotten so much more powerful that you don't need that many members to process the same workload. So they're consolidating.
SPEAKER_01Oh, that's really interesting. Yeah, I hadn't I hadn't thought about it going that way, but I guess isn't that a bit dangerous?
SPEAKER_04Well on a availability point of view.
SPEAKER_02Well, so yes, for for availability, you'll always have at least one more member for availability, but you don't have to have the five, six, and eight week data sharing group that you had before, right? Because you can get a lot more done on one box than than you could previously. And like I said, there are some customers I'm working with that are in the process of basically you know shutting down members in the group because they don't need them anymore. They can accommodate all the workload given you know one or two members instead of five or six. And that's all basically because the machines have gotten so much faster and they have so much more capacity. Well, you know, look at you know, look, look at the buffer pools, right? Remember the days of just three buffer pools?
SPEAKER_01Well, I was a bit before my time, but they they've certainly come on a lot, haven't they, in in the last few versions?
SPEAKER_04In the same way that DB2 didn't stand still and it moved on from 1.1. Uh, so did you. You moved to uh you moved on. So what did you do next?
SPEAKER_02Uh what did I do next? So um so this was at the phone company, and it was quite quite the learning experience. Um and this was also at the same time, I forgot the gentleman's name that predicted the end of the mainframe. Um so my manager came to me and said, Michael, you know, this DB2 stuff, it's great, but you know, we don't think it's the future for us. Um maybe you should learn Oracle or or SQL Server or something like that. And I said, uh I kind of like DB2 on the mainframe. You know, I'm a mainframe guy, I've always been a mainframe guy. Learning this distributed stuff just just didn't. So by happenstance, I happened to have a chat with uh our cat manager firm BMC and started a discussion, and lo and behold, next thing you know, I'm I'm a software consultant for BMC Software, and I've been doing that now for geez, quite a bit of time as well. So yeah, so now it's still working with DB2 on Z, um, but in a different capacity in the sense that I help our customers, as you know, Marcus, basically get the most from our technology and as well utilize DB2 as effectively as possible. A lot of times our customers uh you know don't know as much, uh, so we're you know we try and help them as best we can, even when it's not product related. So that's how I spend my life these days is just helping customers uh effectively utilize DB2 and our solutions.
SPEAKER_01So you've been with BMC for what, nearly 30 years? Is that right?
SPEAKER_02Uh so that's my goal. That's my goal. I'm a little shy of 30.
SPEAKER_00Right.
SPEAKER_02I don't know why 30 has a ring to it, but uh so my manager keeps asking me, well, how much longer are you gonna be? How much longer? I I don't tell anybody, so oh yeah, but now I just said it. It's recorded. But that's my plan. I'd like it's out there now. Yeah, yeah, it's out there. I'd like to hang out for a couple of months, or actually even longer. I mean, the I really enjoy the job. I really enjoy working with customers. You're always learning something as far as the technology is concerned. So as far as I can I can still do the job, and you know, I keep going forever. Theoretically.
SPEAKER_01And we love seeing you at conferences. You're you're a great speaker. You're you're so exciting to listen to.
SPEAKER_02And you've got discussions.
SPEAKER_01But it's not just speaking that you you do with iDug, is it? In the past you were the president, I think.
SPEAKER_02Actually, you're one of the few people that actually even remembers that. Yeah, yeah. Uh so I yeah, I was president for a very short period of time because I think I was elected in May, if I remember. Uh, and then I started with BMC. So the the policy back then was anytime there's a job change, you have to tender your resignation, the board reviews it. Um so we went through that process. Uh and it was um it was quite the tug of war, but ultimately the board decided that no, I had to step down as president, then someone else took over. And that was because I think it came down because I I was a vendor and you know it it wouldn't look nice for Yeah, not completely neutral.
SPEAKER_01I get it.
SPEAKER_02Exactly, yeah.
SPEAKER_01Yeah. But but you have you've always been involved with with iDug on this thing.
SPEAKER_02Oh yes, yeah. I think uh I've gone to every single iDug in North America uh ever since the beginning. There may have been one or two that I missed uh you know for for a variety of reasons. And yeah, and you know, I've gone actually thanks to iDug, I've been able to see quite a bit of the world. I have you know, being on the board, I actually got to go to Australia a couple of times, so that was also very interesting. Oh, nice.
SPEAKER_01Yeah, lucky on my list, but yeah, yeah, it's a long way away, isn't it?
SPEAKER_02Yeah, so it was fun, it was fun going from being an attendee at iDug uh and then being on the conference planning committee, which is where I started, and then you know, moving up the ranks to the board. And see, and this was in the early days of iDug, right? When we used to have the vendor wars and you know, platinum, BMC candle going at it, who could outdo each other. It was quite it was quite the show.
SPEAKER_01With the parties.
SPEAKER_02The par yeah, the parties, yeah. And they were parties. Yeah, it was quite it was quite the show. Uh and like everybody trying to outdo each other, who could throw the biggest party, who could give away the biggest gifts. Yeah. Those are the days. Those were the days, yeah. Yeah.
SPEAKER_01I look back a bit these days, but it's still it's still great, isn't it? And iDugs are such good events, aren't they? They're so well organized, and content is brilliant. And I think meeting people in person and speaking to people is it's just invaluable, isn't it?
SPEAKER_02It is, yeah. And actually, if if you think about it, the i iDug organization has been around for quite a while too. So I'm glad to see that it's still doing well, thriving and surviving in this day and age.
SPEAKER_04Yeah, and it's not just networking that iDug's famous for, it's the actual technical content. Oh, yeah. Yeah. Uh the the speakers, like world-class DB2 experts, year after year. Uh the training opportunities provided are just immense.
SPEAKER_02Right. And I think that that's what makes iDug unique is you know, a lot of the content is from actual users, practitioners, uh, as opposed to the vendors, right? So that that always has a lot more appeal. Uh because it's it's practitioners. So practitioners learning from practitioners.
SPEAKER_01Yeah. But um you you will be speaking at iDug in North America. Um, and what's your session?
SPEAKER_02Uh so that one that one is gonna be, yes, I'll be speaking, but I'll be doing uh an update on our products as opposed to just a general presentation. I okay. I didn't have the time to come up with a general abstract.
SPEAKER_00Yeah.
SPEAKER_02Which reminds me, I need I need to come up with a some for for uh GSE UK.
SPEAKER_01Oh, yes, please.
SPEAKER_02Yeah, they just opened up the call for papers on that.
SPEAKER_01Yeah, yeah, it'd be great to have you there.
SPEAKER_02What makes you think we'll accept you? Well, no. Well, the process is you have to submit an abstract, right? Absolutely, yeah. Yeah, yeah, exactly. So yeah. Step one, submit an abstract.
SPEAKER_04So that after that we'll see. One of the things that makes you popular is the the fact that you're quite unassuming and tongue in cheek. And uh it's like, hey, this presentation is stuff that I wish I'd remembered when. Oh, yeah.
SPEAKER_01Those ones are really good. And that's so useful when you you're just looking for something and then you come across one of Michael's presentations and you're like, oh, and it's so clear and it's all really obvious now. Um yeah, those those those ones are great.
SPEAKER_04He burnt his fingers on this so that I don't have to. Yeah. Exactly.
SPEAKER_01Yeah. And um, Michael, you're based in Canada, is that right?
SPEAKER_02You correct, correct. I live in Montreal. No, well, so I am a Canadian citizen, but uh uh I am uh a native Italian. I was born in Italy, uh the southern part of Italy, and we emigrated here way back in '64. Uh and and uh again, one of those other things, you know, I'm I'm a little kid waking up in the middle of the night in southern Italy. Uh we get on a boat, come here to Montreal. Well, actually, we landed in Halifax, and then we get on a train to Montreal, and we you know, we finally get to the apartment that they had rented for us, and I get out of the car and I see all this white stuff on the ground. I'm saying, what is this stuff?
SPEAKER_00Oh, wow.
SPEAKER_02Yeah, because being from southern Italy, I had never seen snow, and it's like frigid cold and you know, yeah, three feet of this white stuff that you have to wade through. It's like, why did we do this? Yeah, so 64 years, yeah, yeah. Whatever how many years later we're still here. And now my kids are asking, well, why are we here? Why aren't we in California or something like that?
SPEAKER_04Why did you leave it today?
SPEAKER_02Yeah, yeah. Well, actually, they're even upset with me because they're trying to reclaim their Italian citizenship. And they said, you know, because I became a Canadian citizen and I'm not sure, but I think part of the process is you're you know, I had to renounce my Italian citizenship. I wasn't even aware if I did or didn't. Anyway, so they're giving me grief now. Why did you do that? You know, I could have been an Italian citizen now.
SPEAKER_01Well, Marcus and I were debating whether we would be allowed to, like, if it's politically correct to comment on how wonderful your accent is, because you sound like a a classic kind of New York Italian, and I don't think it's correct to say that, but we absolutely love the way that you speak and your accent.
SPEAKER_02Oh, really? Okay. So the funny thing there is I wasn't even aware I had an accent, which is really which is which is really yeah, which is really ironic in a sense, because uh I as you said we just confirmed, I live in Montreal, and Montreal is predominantly French speaking.
SPEAKER_01So do you speak French as well then?
SPEAKER_02No, not that much. Uh because I went I went through school in the English system and work was predominantly in English, right? So the little bit of French I know is what I picked up in high school, which is very, you know, basic and preliminary. Yeah, so no French accent.
SPEAKER_00Right.
SPEAKER_04So you've you're a an established speaker and consultant. Thank you for the call, the the compli wouldn't say established, but thank you. You are you are so w what do you do when you put your feet up?
SPEAKER_02Well, no, so I rarely put my feet up, but when I do, uh I I watch a little bit of television. But uh the other time it's I'm always up to something, like helping my daughter renovate her house and doing a number on my fingers. But uh so my wife and I have a you know like a house so that that always needs tending, and we have three kids, so they're always up and about and keeping us busy. And now we have a granddaughter and she's she's a handful too.
SPEAKER_01So wow, oh lovely.
SPEAKER_02Yeah.
SPEAKER_01How old is she?
SPEAKER_02She's gonna be four years old this June, so we're looking forward to celebrating.
SPEAKER_01Oh, fantastic.
SPEAKER_02Oh wonderful age.
SPEAKER_01Keeps you busy and keeps you young, I guess.
SPEAKER_02Oh yeah, yeah, for sure, for sure. And then you know, there's the occasional every once in a while I get on a fitness mood and I'll go bike riding, but that doesn't last long.
SPEAKER_01I know the feeling. I have the same problem.
SPEAKER_04So I guess you've seen in your years as a software consultant, um, plenty of customers and plenty of uh strange things that customers get up to when they're talking about DB2. What kind of uh horror stories do you think that you could share with our public?
SPEAKER_02Oh, um you know, as funny as that may sound, well, there's a couple, but I'd yeah, I'd rather not get into the details. Uh but yeah, with with customers, it's um it's all over the map, right? Um and sometimes you're asking yourself, well, why are you doing this? And it always comes back because that's the way we've always done it, which I'm sure a lot of us have heard. Uh but specifically to DB2, I'm I'm struggling with that one, Marcus. That's that's a good question. That there's nothing comes to mind right now that that that's really out there.
SPEAKER_01Well, that's good. In a way, that's good.
SPEAKER_04So what would you think would be your favorite or not necessarily your favorite uh functionality in db2 or improvement or innovation, but what do you think is uh has contributed most to business?
SPEAKER_02So there again, uh I have to go back to some of the stuff that IBM's done in the early days, like for example, the Visual Explain. Um, you know, going back to you need to develop efficient SQL, and a lot of times application developers don't know what they're doing. So going back to your earlier question, Marcus, about some of the horror stories you've seen on customers, some of the SQL I've seen out there, it's like, well, why did you do that? Right? Like Cartesian joins, it's like Cartesian products, right? Like, why? So the yeah. Uh so yeah, visual explain. Um I personally liked again going back to you know the the the analogy we had with packages, um the case logic, right? Being a cobalt programmer, being able to do case SQL statements, right? Stuff like that. Uh and then I, you know, obviously closer to today, uh that continuous delivery thing, I think that was, although it was kind of painful, I think, at the beginning, I think was a really good idea on IBM's part. Uh and it it also uh not necessarily forced, but compelled the vendors to go along with the same suit. So uh it's making it a lot easier for customers now to adopt new technology a lot faster rather than waiting two or three years for that big release and then being able to exploit something.
SPEAKER_01Yeah. And uh your your views on all of the AI stuff that we've got available and that's that's coming these days.
SPEAKER_02So that that is Um I'm on the well, I'm not on the fence. Um I'm trying to see how it's gonna affect us. I'm trying to see how best to deploy it. Uh it it it's it's it's a new field uh for me specifically. So I'm just trying to wrap my head around the whole thing, basically.
SPEAKER_01Yeah, and there's there is so much to consider, isn't there?
SPEAKER_02Yeah, yeah, yeah.
SPEAKER_01Um I think finding the use cases is still a bit of a challenge, isn't it? Particularly with big db2 customers. Um it it can be quite hard to to think of exactly how it can be exploited in the best way. But yeah, it's interesting. And uh outside of db2, I'm I'm quite I quite like some of the things that that we can do with with AI, but when it comes to DB2, it's I think it's a bit more tricky.
SPEAKER_02And and yeah, and there's a lot more factors that you need to to to look at. So and we were we're doing some of that at BMC as well, in the sense of trying to infuse AI in a lot of our products. So it it it's yeah, I it's gonna be a challenging field and it'll be interesting to see how we apply it effectively, that is.
SPEAKER_01Yeah, it's certainly gonna keep everyone on their toes for quite a few years, isn't it?
SPEAKER_02And and basically, you know, trying to come up with ideas as you know, like what can you do to exploit AI, right?
SPEAKER_01Yeah. Yeah. Yeah. What's that what's that thing that magical thing. Yeah.
SPEAKER_04Silver bullet. Yeah, exactly.
SPEAKER_01We we've obviously spoken to quite a few people over the last couple of years that we've been doing these podcasts. And we we do get kind of people love it and hate it all at the same time. So yeah, you're not alone.
SPEAKER_02Well I wouldn't say it's the hate part, I'm just nervous it might uh be either overused or not applied correctly or I think it needs to be carefully thought out. And especially in some fields like you know when we when we talk about using AI in the field of medicine, there's no way I want an AI to tell me here are my symptoms, oh this is what you've got.
SPEAKER_00Sorry Yeah it's fine for rewriting emails but yeah please don't diagonize me.
SPEAKER_04Yeah I was talking to someone earlier about AI and I mentioned that I worked in the mainframe techno space and she said so how does AI work on the mainframe? You know I thought the mainframe was like a legacy system just running old COBOL programs for banks that uh Elon Musk and the Doge are trying to get rid of by replacing them with dynamic Java. How does AI sit with the mainframe? So I had to explain you know what the Z17 was and the onboard inference engine and the telem chip and the Spire processes and whatnot.
SPEAKER_02And they were like really but but but that's another interesting point though Marcus is uh you know people look at AI and what scares me is well AI is just code right at the end of the day it's just code. Someone makes a coding error for lack of a better word who knows what's going to come out on the tail end. I mean you know it's obviously it's a lot more sophisticated but at the at the end of the day it's just code.
SPEAKER_04Yeah yeah and uh at the end of the day it it's it's a human that's written the code exactly fed the uh the machine learning algorithm with its large language model or something or other or trained the model with uh with the inherent human biases one way or another that's true yeah the whole thing about vibe coding uh what oh Julia you haven't come across vibe coding I no and neither am I Marcus share share okay so I come across this a couple of weeks ago and uh it was only a couple of weeks old then oh okay I don't think so bad now then so you're not you're not that far behind uh so vibe coding was a a term coined by this expert well respected programmer whereby you don't actually need any skill or training in software engineering whatsoever to write a computer program.
SPEAKER_01You can just rock up with a microphone plug it into your chat GPT and say write me some code which does this on your phone and after a few iterations ta-da it did it wow yeah wow I I don't know if that's wow amazing or or wow scary but wow so that made my blood run cold hearing about that thinking I mean after all of the training that I've had this is horrific it also made my blood boil at the same time thinking where is the skill where is the the acumen that's involved in in crafting code to do what you want where's the testing going on but then I realized actually I'm the kind of person that does this already because hands up if you haven't asked chat GPT a question hey how do I do this?
SPEAKER_04How do I you know what does this mean or help me write this blog post, help me write this computer program. It's great for prototyping in my opinion but I wouldn't let chat GPT written code anywhere near production not without severe peer review and security checking.
SPEAKER_02Oh for sure for sure yeah yeah and to your point testing right absolutely it might be good for useful for prototyping yeah as as a starting point.
SPEAKER_04But it made me think hey maybe you could do vibe DBAing this table space.
SPEAKER_02Yeah well actually there is work uh along those lines right where you know people are like I guess trying to standardize for lack of a better word and so that's kind of like an extension of AI in my opinion but yeah so all kinds of interesting ideas out there.
SPEAKER_01Yeah we live in interesting times is there anything in DB213 that you've been excited by or even DB212 because a lot of people are still using DB212 are there any features from from either of those versions that you've been really excited by I was very excited about SQL data insights but I haven't really had I haven't really had a chance to play with it or to get it to work for me. So my interest in that has kind of gone off a bit uh but there's there's other things which I just think are are brilliant now that that are that have come out recently.
SPEAKER_02To be honest with you Julie that's where I was gonna go is was was this SQL data insights but similar to you I really haven't had a chance to sit down and play with it. But that was real interesting that um and what they've done with DB2 for ZOS AI or whatever so I think that used to be the um the AI for the optimizer and I think IBM's since extended and enhanced that significantly where they're doing a lot more.
SPEAKER_01So that yeah that I I'd love to sit down and get some hands-on experience with that just haven't had a chance yeah it's it's hard to find the time because now there's so much in db2 isn't it that like where where do you start where where's the next what's the next thing that you're gonna be able to learn and yeah finding the time to do that on top of your your day job is quite a challenge.
SPEAKER_02Oh yeah absolutely yeah and I think well also there's in my experience or the customers I work with not a lot of them haven't embraced a lot of this newer stuff so there's less compelling motivation to stay ahead of the curve right yeah yeah that's true.
SPEAKER_01I I know as well that that some places would rather keep things uh very stable and static and not really exploit any of the functionality even from version 11.
SPEAKER_02So yeah yeah and to your point julia the the customers that I work with they're still they're you know they're slowly migrating to DB213 a lot and a lot of them actually a lot in the more in the field are only going to 13 because 12 is going out of support so they're they're legally enticed basically is that is that December this year that it's out of support or is it sooner? Uh December December December of this year I believe it's only seven months then not long is it?
SPEAKER_04Yeah yeah so open your checkbook and fill in extended support here exactly right yeah but I was excited when Hawkin um said that there will be a V next of DB2 and I'm trying to remember whether he did he say it wouldn't be 2026 but maybe 2027 and I was excited about these things even though yeah it it's just hearing about them it gives me a bit of a buzz.
SPEAKER_02It's it's exciting to think that that it's it's just gonna keep going and keep getting better and bigger yeah and the and the stuff that they're IBM's going to do with the engine is it's it's it's mind-boggling. I'm really in awe of those guys in the lab and and the stuff that the ideas and the innovation that they come up with it's fascinating.
SPEAKER_04It really is a clever bunch.
SPEAKER_02Uh-huh uh well I I don't know if it's clever or imaginative I think unless they're all using chat GPT to to do their programming for them.
SPEAKER_04Yeah vibe DB2 what was the the terminology used in version 12 when they started to release functionality in function levels uh so that's been going on since version 12 no wasn't that continuous delivery yeah that's the one yeah yeah um I'd like to know I wonder if there are any metrics out there on how quickly customers adopt the functionality that's delivered in continuous delivery as opposed to in the the good old fashioned versions because I remember when a version of DB2 would come out we'd be like okay what now what can we do? And we'd spend the rest of the lifecycle of that release of DB2 or version of DB2 just trying to backfill all of the business requirements with that function.
SPEAKER_02That's an interesting observation because again given the customers I work with um what I see in the field is a lot of the customers are very aggressive with the maintenance. So the code level of db2 might be at the latest level but if you look at their function level or their Apple Compat it's way back there. I mean I I'm working with some customers that still have the default Apple compat at V10 believe it or not. So yeah yeah yeah so in terms of exploiting the features and functions I think and and that that's probably why IBM did that the continuous delivery thing is so that pe customers can pick and choose what they want, right? And they have more flexibility and more control.
SPEAKER_01Yeah yeah has has there been any word about IBM deprecating Apple Compact V10?
SPEAKER_02No not so far not so far. I imagine at some point they'll get around to it but yeah maybe that will be announced uh this year or next I would suspect definitely if if they announced the V next I would suspect the V10 on the N11 will probably be gone from an Apple compact support perspective but just speculating on my part.
SPEAKER_04Yeah yeah me too I I knew of some customers who were forced to rebind when they went from V10 to V11 that hadn't rebound in decades before that.
SPEAKER_05Yeah.
SPEAKER_01And now they have to do it to get to 13 as well don't they?
SPEAKER_02Yeah. You're right to get to 13 there is there's some rebinds that would have to be done based on the last time you rebound or what version of DB2. I want to say it was version 11 but I'm not sure actually there you go I have to go look at one of my foils on my presentation because I did talk about that yeah because some customers believe it or not to save a few milliseconds of CPU will turn off that that ZPARM that tracks when the package is last used.
SPEAKER_01That's such useful information.
SPEAKER_02It is why would you do that? Yeah because you know DB2 will check when's the last time it was bound. Well yeah and and even if it falls through that loophole uh the structure's still not going to be valid so DB2 is going to tell you hey buddy you got to rebind this guy because I can't execute him so one way or another it'll catch it. It's just it would be nice if you if it was caught beforehand and you had the control and you could rebind rebind at your leisure, right? As opposed to defaulting to auto bind.
unknownYeah.
SPEAKER_04Yeah because as much as we like automation and we want the system whether it's DB2 or ZOS to be self-managing that there there's still an element of control that we would like to exert over our domain.
SPEAKER_01So okay don't don't automatically rebind and and puff when I'm not expecting autobind is one of those things that I've I kind of swing back and forward on whether I like it or not and I s I s I never really know like one minute I'd be like yay autobind is great autobind is fine but then the next minute I'm like is it what what do you think?
SPEAKER_02Well I personally like it uh and as all things in DB2 right there's there's always the one in a minion where if all the stars and everything line up it can cause havoc right but in general it you know it it works great. And I'm sure there are horror stories out there about where Autobine kicked in and then all of a sudden you know the machine went down. But those are you know it's rare right and then again it's it's a situation of where everything had to line up to create that catastrophe which doesn't really all happen all so it's you know it it's do the math type like exactly yeah it's a numbers game isn't it what are the chances 99 times out of out of a hundred it'll work. That one time something went wrong. So but it it's it's the oops that people remember right like oh yeah it's enough to tell me about that what's that saying one oops will wipe out a thousand out of boys type of thing or something like that I haven't heard that before but that's it's that's quite a painful PTSD triggering reminder.
SPEAKER_01Michael have you got any horror stories from your your travels around the world?
SPEAKER_02I do but they're not related to DB2 uh but it it is a fun horror story so this goes back to even before I was a a programmer uh or in db2 dBA I started out work in in my career as a computer operator uh on a honeywell system no less it was still a mainframe but honeywell uh so I was a computer operator uh and in those days uh the machines were you know water cooled and not like today right uh and the air conditioning system malfunctioned so we had uh a little safety valve a thermometer on the wall that basically allowed you to raise the temperature of the room so that the machine wouldn't shut off so the machine you know it was getting warmer and warmer and warmer and it was getting close to the end of the day and the uh the users had to do their last bit of online processing so in an attempt to extend the day I went to change the thermostat raise the temperature but I did it the wrong way so I basically inadvertently shut the whole machine down everything came crashing down phones were ringing off the hook Michael what did you do? What did you do?
SPEAKER_04So I learned the hard way about temperature controls in computer rooms so that that yeah uh that was but DV2 no DV2 I think um I really haven't really well there's always the old you know like we were talking before you about silly SQL codes uh silly SQL coding Cartesian products that kind of stuff but nothing nothing that drastic no no horror stories personally uh there are there are some that customers have experienced um you know like for example I think I recall one customer that lost their DASD where the DB2 logs were on so you know that that created a nasty not a nasty but a very difficult recovery situation uh things of that nature but I personally uh I've been fortunate thank heavens goodness yeah that that does sound uh that would make my blood run cold for sure yeah and this is yeah you know DASD never fails but every once in a while it does go belly up uh and yeah this particular customer they had all their active logs on that that controlled unit yeah right out was not not the nice recovery situation no no I can't even really start to compute that in my head yeah and then and then there's always you know like the the again the silly stories about oh I thought I was in test and someone did a drop right I've had I've had a couple couple of those um but yeah I've I've been guilty of that a few times unfortunately I've had some clumsy moments well as always it's been a pleasure to talk to you again Michael and uh we wish you all the best and we look forward to seeing you in well we won't be there but we'll look forward to hearing reports of uh your presentation in IDUG North America and hopefully see you in IDUGAMIA and GSE UK later on in the year it'll be my pleasure and and and thanks for having me hopefully uh I've given you guys some entertaining stories absolutely it's been brilliant to it's been brilliant to speak to you and uh yeah we've loved having you on.
SPEAKER_01All right take care folks thanks Michael bye visit iDog.org for the latest DB2 community news information about events and all of our technical content tune in again soon for the next podcast at idog.org forward slash podcast