OpsCamp Debriefing – IT Management & Cloud Podcast #66

After the fun and successful OpsCamp Austin 2010, John, myself, and several others chime in on how it went and what we talked about. We recorded this in the bar/coffeshop next door, so it’s a bit loud, but the recording is nonetheless understandable.

Download the episode directly right here, subscribe to the feed in iTunes or other podcatcher to have episodes downloaded automatically, or just click play below to listen to it right here:

Thanks to everyone – organizers and attendees – I had a tremendous amount of fun and it was professionally helpful.

Check out John’s post on the topic for more of an overview – and we’ll see what the transcription service can make of this one 😉

Full Transcript

(As this was in a loud room, the transcription probably has some errors, including lots of “Male Speaker.” Be sure to listen to the original audio if you think we’re saying something overly weird. -Coté.)

Michael Coté: Well, hello everybody! It’s the 30th of January, 2009, and this is after — what are we here for?

Male Speaker: OpsCamp.

Michael Coté: OpsCamp, down here in Austin. This is a special IT Management, having some beers edition. And this is one of your co-host, Michael Coté, and of course we are joined by John Willis.

John Willis: johnmwillis.com.

Michael Coté: That’s right. And let me take these earphones out of my ears so I stop freaking myself out.

So John, what did you think of the conference today? Did it accomplish the goals you had personally?

John Willis: I honestly think it did, I really did Michael. I think that as we sat down and we thought about this, there are a couple of things we wanted to accomplish. A, we wanted to inform the industry that there is this special type of thing, infrastructure as code, some people call it automated structure.

Michael Coté: Very automated, something or another.

John Willis: And some people call it, what do you guys call it man? You have a new name for it, this really cool thing.

Male Speaker: The model is the management.

Matt Ray: Dynamic infrastructure management.

Michael Coté: That’s Matt Ray.

John Willis: The famous Matt Ray from wacky predictions.

Michael Coté: That’s right. Also known as male speaker.

John Willis: Male Speaker of interest.

Matt Ray: But no, I am serious, I really am.

Michael Coté: I have to say, I think you are pretty good at like helping industry run this stuff there. Like I was — now that I do several of these things, I was paying attention to you and Dave, the other guy, and yeah, you really have it down well.

Matt Ray: I was thinking, ‘Wayne’s World’, did I suck? So I guess I didn’t suck.

Male Speaker: I hear the Emmys might be looking for a host like that.

Michael Coté: There you go.

Male Speaker: Well, you know what I was impressed with, like what I was talking about earlier is that, I think there was actually a pretty good mix of buyers and sellers essentially, of like vendors and actual IT people, which, even at some of the Cloud Camps I have been to, is not always the case, it’s usually kind of like vendor heavy. So it was actually really fun to hear from both sides, and to kind of see the interplay between the two sides.

Male Speaker: Well, there was a lot of passion, I mean honestly, there was a lot of — it was interesting. So today at one point I was like kind of amazed at, we are talking about the stuff I love on a Saturday afternoon. I mean, the shortest delivery, or management of monitoring, or configuration of provisioning, and here we are a bunch of geeks, which is cool. I mean, that shows that there is — I have always thought, I love this, so I have a career, but we got like 60 — I mean, I don’t know what the headcount was today.

Michael Coté: Yeah, we will have to check what it was. It was actually really impressive. I was looking at the — I had been monitoring the event sign up, which I think is always cool, like I was saying, because you can see who signed up in. I think it was like 20 people for a long time, and then it suddenly spiked through a bunch of people. So that was actually pretty nice.

Male Speaker: I think a 130 registered, I think we probably got 75 people here.

Michael Coté: I mean, so as far as the top like — I remember the topics being — like the things I went to, I think you and I only went to one or two — one thing that was the same, which is excellent coordination.

Male Speaker: You are trying to cover the whole —

Michael Coté: The very thing. So I went to the evolution of the — or Agent Evolution, which I thought was pretty fun. I mean, that ended up basically being, what are we going to monitor in the cloud, and there was a little bit of, what is the architecture of agents need to be to fit into a cloud world, if you will, and that was kind of enjoyable.

And there was kind of the ongoing doke in that session — the ongoing doke, it happened once or twice, was, that essentially like these heavyweight agents that everyone hates from the 90s, are kind of like seen to be what you need to do cloud stuff, because cloud stuff doesn’t monitor itself, kind of like infrastructure does nowadays, so you need agents to do a lot of that work, or something along those lines. And so, I think that was kind of interesting.

And the other takeaway that, I think I was saying this, so maybe it’s just — I remember only what I said, which is quite difficult. But it seems like the agents that you build have to be as equally sloppy and non-reliable with the cloud stuff that they are monitoring.

So when you have behind the firewall infrastructure, it’s more or less reliable and you know about it, so your agent can kind of act accordingly. But if you have an agent that’s monitoring some cloud stuff, it’s probably going to be up and down, you are going to lose connectivity and all sorts of crap and the transient thing, so your agent has to act accordingly. But that’s what I remember from that session.

Male Speaker: Well, for me, I wasn’t in that session, but I definitely liked GroundWorks now, because Tara, she hugged me, so it’s the highlight of my day, really, so I am really good.

Michael Coté: You meet nice people, you have nice thoughts, that’s the way it goes.

Male Speaker: So I think while you were in there, I was in a service delivery.

Michael Coté: How was that?

Male Speaker: It was pretty cool. I mean, we talked pretty much about — I tried to figure out what configuration was and what it means in the cloud and delivery of — so we had a lot of discussions about, the Amazon has set — really caught us all off guard.

Michael Coté: Right, right. Matt Ray was talking about that.

Male Speaker: Right. We are all thinking about infrastructure, most of us are doing it for many years; doing monitoring configuration provisioning. And then all of a sudden Amazon kind of comes out and says, this is what we have, and so for about a year-and-a-half or two years, we are trying to figure out, first, what is this crazy thing, and now we are looking at it and saying, holy crap, we have got to be like that.

And I think even the guys in IBM, Fritz was saying like, they have kind of put us in a position where we all have to now say, we suck at infrastructure. I mean, we don’t know exactly how they did it, but they are doing something great, and we got caught with our clothes off, and we have got to basically try to figure out how to get to that spot.

Michael Coté: Well, interjecting the conclusion, I would have a day in the middle, I think that’s the main conclusion I came to is, I think this bucket of thought, if you will — like I think the thing it really needs to do is define like — it needs to have this statement that like, that stuff in the past has broken and no longer works, and here is why our stuff works. Like here is why this stuff is like the better way of doing this, because I think it’s a little unclear like what — maybe the Cloud stuff is cheaper or whatever, but there needs to be something out of it.

Male Speaker: It’s not cheaper. I mean, I keep hearing that, and we keep circling back to that, but the reason Amazon works is because there are zero people involved with the transaction. It’s not the $0.10 an hour. I mean, that’s great, and it sounds good, and it makes sense to a point, but even if you add up the numbers, if you are running 24/7, 365, but the thing that really — I don’t think most people get why that model of IT infrastructure from a service delivery standpoint is so great, is because it is a zero touch transaction. You don’t have to deal with a person to get what you want.

Michael Coté: What were you going to say Matt Ray?

Matt Ray: I mean, zero touch, that’s awesome, but what I was hearing that really changed the game was the fact that, we no longer have to care about root cause analysis. It’s like, throw the box away, give me a new one.

Someone said, we have gone to the Windows model. It’s like, something wrong with the server, reboot it, toss it out the window, give me a new one. We are just failing forward and rolling out new infrastructure to just replace whatever was wrong.

Male Speaker: I think as John was — I mean, that’s the kind of like — there is nuance to that, but I mean, that’s the kind of like statement that I like coming out of an event like this. It’s kind of like — during today I have been thinking about, like when I was observing the Agile software developers, like kind of working out the industry, and it was full of stuff like that.

Male Speaker: Agile drinking technology.

Male Speaker: I mean, it was full of like all these statements that weren’t absolute statements, that really weren’t absolute statement, but it was more like, this is like what we are aspiring to, and kind of what we are trying to drive forth, and that’s like one of it, is in reality you are going to do some root cause analysis but —

Male Speaker: I don’t think that will ever happen, I mean honestly, I really don’t.

Male Speaker: Just like with Agile, you are never going to have 100% 8:18 that work all the time.

Male Speaker: We have always been hearing there will come a day when we won’t have to do that, and the reality is, we never get there, and that’s okay.

Male Speaker: But you will do less of it.

Male Speaker: Well, we have to understand the shift of how we do it or what becomes important, because things change. We have the luxury of having a lot more leeway about having to care.

Male Speaker: Or we can care about more important things.

Male Speaker: We can care about hiring the staff.

Male Speaker: Exactly, exactly.

Male Speaker: Hopefully, we don’t have to worry, because maybe we can — if we do —

Male Speaker: The Alabama, Texas game?

Male Speaker: What?

Male Speaker: I am sorry.

Michael Coté: A good segue from that is, for the second session I was there, was the — what was it, like the craft of being a systems administrator or something, which I thought was going to be like the usual sort of boondoggle of like philosophizing, but it was interesting to hear — like people were very involved and saying a bunch of interesting things. It was interesting to see people almost get to the point of figuring out, so if I am a [?], or I am selling [?], and then everyone uses Google Apps, what does that mean? I mean, to put it in the extremes right away.

Male Speaker: How can I make it simple enough for —

Michael Coté: Right. And then it did sort of — halfway through the session it didn’t evolve; that’s the wrong word, but it became a BSM discussion, that was basically like, why does everyone hate IT so much, and like, what can IT do to make them like us better? And then to short change it, basically you need to figure out some way to tell the business, you are helping them make or save money, which I think is always a healthy thing to always be aware of, like you always need. So that was nice.

But like there was a moment there where it was getting to that point of, what’s going to be the future of IT if all this cloud hoopla actually works out only half of the way? I think that’s a discussion I haven’t really seen IT people having, so much as developers having.

Male Speaker: In the parallel session, which was Service Assurance, we were taking about what IT did, what can we do. Because you have got this IT infrastructure where parts of it are not local. There is cloud, there are services, there are things. What can you do to avoid downtime when your transaction processor, who is not on site, who you don’t have any SLAs from, how do you handle that, how do you monitor that, and what do you do to prepare for it?

And people are like, well, I have got black holes, I have got big blind spots in my model. I don’t know what my network looks like. I don’t know what my services look like. For solutions to work well, we need better modeling, we need more synthetic transactions. We need to know where our blind spots are, so at least when these things go down, we have a quicker storage. I mean, there weren’t any great answers. Eventually it turned into a Zenoss sales pitch, so it all ended up well. We got a lot of converts, but still. Nobody had any great answers.

Michael Coté: Yeah. Like during that — on that topic, the thing I didn’t pipe up about was, system and crafting. And I was talking with Erik, CTO of Zenoss about this, like afterwards, is that, it seems like the interesting opportunity for operations people with all this cloud stuff is to actually have some thing they can bring to the table, that is something new, that can generate revenue. Like development can always bring something that helps generate revenue.

But like IT is just sort of like, it’s like a bummer. So like, it seems like, IT could say like, hey, there is this new way of running our business that may well have different thing that we can sell or different way we can make money, and that’s kind of another discussion it would be interesting to see like IT people have. Like what’s the revenue generating brand new innovation to the business that you can bring because of all this cloud stuff?

I don’t really know what that is, but that seems like an opportunity that’s there.

Male Speaker: Well, some of the other things. I think kind of early on we started out — we ran this thing as kind of a classic cloud [?], we did the lightning talk, and then we did the [?] session, and we invite people up. And we were kind of starting to get top heavy on configuration management. So I just happened to accidentally mention the word monitoring, and then the conference really became monitoring camp.

Michael Coté: That’s an interesting like division that happened, not division as in people — a separation of concern, to use some crazy phrase.

Male Speaker: I don’t remember who had the slide, but someone had a slide that on one side was management and the other side was monitoring, and it’s like, how do we bridge that gap? How do we tie those together?

Michael Coté: So what was the third and final section that you guys went to?

Male Speaker: Well, there was a sales pitch session where all the vendors — but none of us sat in on that. The session that the rest of us sat in on was

Male Speaker: Oh yeah, we were all in there.

Male Speaker: We were all in there. So there was a great session from IBM. Phil, do you want to tell us about your session? You want to introduce yourself.

Phil Fritz: Yeah. Well, I am Phil Fritz, I work for IBM. We had a session about monitoring from cloud. It was a pretty interesting conversation. Well basically, I shared a couple of initiatives we had about putting up our IT and monitoring concept on the Amazon Cloud, so folks can monitor there. You see two instances from the cloud.

Michael Coté: So what was like the good insights you got from that?

Phil Fritz: Well, there were a lot of questions about what people were using cloud for the required monitoring. There is still a — they see a lot of development tests, we use [?] for development test, but who monitors that, or how many production systems are actually on Amazon.

And for the types of folks that are doing production on Amazon tend to use smaller Web 2.0, Facebook application, guides, other pieces, they tend to use basic tools.

Male Speaker: They use a lot of IBM too.

Phil Fritz: Oh yeah. Either go without, or they just use CloudWatch or they use Open Source and things like that.

Then we also talked about the other offering, IBM provided cloud service for monitoring.

Michael Coté: Right, right, right, which is available as a video overview and demo.

Male Speaker: At redmonk.com.

Michael Coté: That’s right.

Phil Fritz: But yeah, what was interesting to hear was, the trail of dead that SaaS and systems management has left, and many have tried and died, so it’s sort of the challenge in the marketplace of, providing service and systems management tools out of the service.

Michael Coté: So were people positive about SaaS based IT management or were they negative about it?

Phil Fritz: I think optimism always proves eternal, because —

Michael Coté: I mean, that’s kind of a change from like how it was recently. I think here people like actually think like SaaS based — like something like Tivoli Live and others to work is somewhat of a change. Most people are pessimistic about that.

Phil Fritz: Well, I mean, I think, at the end of the day the model is compelling. It makes too much sense for it not to work. I mean, someone has got to figure it out. So like at the end of the day — you have got experience with this, right, you have done some of that stuff in the past.

Michael Coté: We all have.

Phil Fritz: So I think at the end of the day it’s like, it makes a lot of sense, someone has got to figure it out. You can argue there are players in the field that can figure parts of it out.

Michael Coté: So it sounds like it was overall positive for you then?

Male Speaker: We got another celebrity who just came up.

Phil Fritz: Oh yeah. I was able to spill a lot of IBM secrets for you. Those guys were furiously taking notes.

Michael Coté: Here we are again, it’s the bar, it’s after hours, in the IT Management Podcast. We have another celebrity. Come on here Damon. Damon from DTO Solutions.

Male Speaker: You can uncomfortably bend over.

Damon Edwards: There is something I always wanted to say, well, hello there! This is the IT Management Podcast, I am one of your – am I a host, am I a co-host? Oh, hell, I am Michael Coté, where is my beer?

Male Speaker: You are that John M Willis guy.

Damon Edwards: That guy?

Male Speaker: He is a nob.

Male Speaker: John M Willis, at johnmwillis.com.

Male Speaker: John M Willis should have stuck to his —

Male Speaker: Sometimes John Willis at johnmwillis.com.

Damon Edwards: Nice, nice.

Male Speaker: That ends the comedy hour.

Male Speaker: So [?], I mean you did a lot of —

Damon Edwards: Yeah. I think the concept, next generation, is one of the most overused, kind of burnt out, terms in technology. I think when someone hears next generation, they probably think, oh, it’s 20% or 30% chance it’s going to come true.

So I think us running around saying — yeah, we are going around saying, we are the next generation of IT operations or systems management or whatever you want to call, IT operation management, we are the next generation.

When we say that collectively, people just kind of think that we are crazy, we have got some newfangled ideas. And I think we live in kind of an echo chamber where we are doing some amazing work for the old companies, big companies, people make a lot of money on the backs of the work that we are collectively doing.

And I think that a lot about today was kind of coming to terms with that, and trying to figure out how to put the best foot forward. So as an industry and as a group we can do a better job of promoting our successes and really explaining the vision of what — not the vision, the reality of what we are doing.

So when we say next generation, we are actually delivering on that promise. And I don’t mean this as DTO Solutions or ControlTier, I mean this as a group across the board, Puppet Chef and ControlTier, you name it, we are making this stuff a reality, and I think we need to do a better job of telling that story to people.

Michael Coté: No. I mean, that’s a good point that I was thinking when John was saying something the other day, that the — I probably spend a lot of hours trying to think of a better phrase than next generation, because like it happens a lot where someone is like, hey, I have got a new way of doing something, and like I can’t use all these great phrases we use over the past 20 years, because people kick me out of the room if I use it.

So you have got to come up with something new, and I think that is — one of the exciting thing for IT about cloud stuff is that, it hasn’t been kicked out of the room quite yet. So people, whether they — for people who have something that’s even kind of cloud influenced, if there is actually a new, better technology, there is this window of opportunity where people might actually believe you, which is, it doesn’t really ever happen anymore in IT at all. So it’s kind of refreshing that you might actually believe it.

Like I mean the last time that happened was like with Open Source, I think, and then we got burned by SOA, and we are still kind of recovering ourselves.

Male Speaker: The trough of disillusionment.

Michael Coté: That’s right, but it never got out of the trough, that’s always the problem.

Damon Edwards: Yeah, and I think that’s — I think that’s a challenge that we face, is to show the world that it is a reality, and come up with some common messaging. And I don’t mean that in the bad marketing sense, I mean just sort of in the vocabulary sense, to help users.

Michael Coté: And the final section, the one that you didn’t see or whatever, was sort of figuring out the automation tool change. I mean, I guess I was trying to encourage you all to sort of like figure out what — explaining to everyone like what exactly — like why should I have one?

Damon Edwards: I can’t tell you how many times we go in to talk to CIOs, or even people down in the trenches, they are just confused by all the different tools out there, concepts, and they don’t realize that they all sort of fit together and do good things. It’s just a big confusing mess.

So I think a lot of this is about trying to reduce that confusion of users, and it helps the industry as a whole enable to do that and drive that conversation.

Michael Coté: Definitely.

Male Speaker: The entire world is meant to come up with the new [?].

Male Speaker: I think at the end of the day, the idea here for OpsCamp was to find out if we had enough people interested in talking about something we are all passionate about, and I think today was a huge success.

Michael Coté: Yeah, yeah. To see if there was enough like — if the cloud gave enough grease to the wheels.

Damon Edwards: Yeah, and not just the cloud. I mean, we talked a lot earlier on about a cloudy world. It’s not cloudy, it’s virtualization, it’s actually folks who are going to have to deal with their [?] but we are ops people, where do we fit in? It’s crazy out there.

Michael Coté: Well, I think that’s a good conclusion.

Male Speaker: Oh, wait a minute. Has everyone seen the downfall, the Gartner bid, with the downfall of Mashup, just had to bring it up, where they were talking about the AR analysts.

Michael Coté: They are a challenger, but not in the leadership offering.

Male Speaker: Make sure it’s in the links. Make sure you have seen it, because there is a shout out for RedMonk.

Michael Coté: Yeah. The one person you wouldn’t want a shout out from.

With that, we will see everyone next time.

Disclosure: Groundwork Open Source, Zenoss, IBM, Spiceworks, and several other people mentioned or related to these conference are clients. See the RedMonk client list.