DevOps Chat: Purpose Built Appliances for Containerized Applications with Diamanti’s Mark Balch
DevOps Chat: Developers and Security with Pete Chestna, Veracode
Patrick McClory on Building Successful DevOps Organizations

Shimel: Sure. Cliff, what do you think about that?

Utstein: Yeah. As CJ said, a lot of organizations right now are looking to build cognitive applications and what they’re thinking about often, with that, is how to move to the cloud, how to improve their overall processes within their organization, and even how to change their culture. And DevOps cuts into all of these—giving an organization a way to go through, really, a set of changes that let them develop a much tighter customer relationship, ultimately, through faster iterations in the development cycle, getting to the customer with their MBT faster, and doing it in the cloud so that they’re able to utilize all the advantages that the cloud offers them, as far as getting to the customer faster and understanding what the customer wants.

And what’s behind this, as CJ said, often, is a whole set of cognitive needs to improve their apps, based on utilizing these cognitive services, so we see it in the financial services space; we see it in health care; we see it in retail. We see it in a lot of industries, where businesses are looking to incorporate these cognitive abilities into their apps, to build the customer relationship, ultimately.

Shimel: Sure. Hey, Cliff, if you don’t mind, I wanna dig in a little bit on the whole cognitive thing. You know, with IBM, a lot of people sort of make Watson and cognitive synonymous, right? I mean, it’s the IBM brand, if you will, for cognitive, if we wanna say that. But we’re hearing so much today about artificial intelligence, machine learning, right, automation, all of these things. And automation is obviously closely tied into DevOps. Just to be clear, Cliff, when you’re talking about cognitive, first of all, it is synonymous with Watson and certainly for IBM’s purposes. Is that correct or incorrect?

Utstein: You know, I think that Watson provides us with a whole set of tools and applications providing cognitive abilities. I think, in addition to that, however, we can look at the cognitive abilities in DevOps as well, in machine learning, to determine some basic things like whether a release is likely to succeed or fail in production. We’ve got algorithms, cognitive algorithms, around that today, to help organizations figure out whether or not they’re gonna have success with a release, based on prior releases, based on patterns that we’re able to interpret within the code as well. So it means a lot of different things, Alan.

Shimel: I agree. Agreed. I just wanted to clear that up for our audience in there too and I think, you know, and, in my mind, anyway, when I look at the relationship between DevOps and cognitive and cognitive and DevOps, there’s a little bit of “chicken and egg,” in that I think cognitive can help you with your DevOps transformation and move to the cloud and everything that’s wrapped around digital transformation, while, at the same time, that same digital transformation and DevOps transformation can help you, you know, accelerate use of cognitive technologies to make your business better. So it very much is a win-win, two-way street. CJ, what do you think?

Paul: Yes. Absolutely, yeah. One is actually the building and deploying of cognitive applications itself and applying the DevOps principles; the other is using cognitive technologies within the DevOps tools themselves. Yeah. They’re both very relevant and things that IBM’s been working on.

Shimel: Excellent. Guys, if you don’t mind, ’cause time is always an issue here, I’d like to move on to the next topic, which is blockchain. So, you know, I just came off of a world tour roadshow around DevOps and other kind of issues, and, wherever we went, blockchain was incredibly hot, was incredibly popular. People wanted to know about it; people wanted to hear what the story was around blockchain. But, at the same time, I had people who were like, “What is blockchain?” Right? They didn’t understand what we meant by “blockchain” and why was it tied into bitcoin and these kinds of things and why should we care. Cliff, if you don’t mind—and I don’t know how much you know about it and, if not, CJ—can we do a level set for the audience about what we mean when we say “blockchain”?

Utstein: So, you know, I think that might be a different talk, Alan. I think, from a DevOps perspective, what we’re seeing is the need to deliver the processes, the culture changes, the tools to support developers that—especially in financial organizations, that are today looking to incorporate blockchain into their applications. And that’s really been our focus, is helping them do that quickly, securely, with governance and all the controls that a financial organization needs in order to be successful with blockchain.

Shimel: Good. Fair enough. I think that’s kind of DevOps  CJ, what do you think?

Paul: Yeah. So blockchain is the Hyperledger technology that underlies offerings such as bitcoin, but that underlying technology, though, can be used to secure financial transactions, right, or different kinds of transactions around the world. So it’s driving a level of transformation in the financial industries, that lot of companies wanting to build new applications to take advantage of this technology. So that’s where, as they try to build these new applications, the DevOps techniques and tools that we’re providing can greatly help them, you know, build those applications and get those applications out there faster. So, essentially, DevOps is sort of the foundation that helps them innovate more rapidly in these new domains.

Shimel: Got it. So, Cliff, you’re the product guy; I wanted to talk a little—ou know, what are IBM’s customers doing around DevOps and blockchain? Can you give us a couple of examples? And you don’t have to mention names—you don’t have to name names, obviously—but couple of examples.

Utstein: Yeah, I think—I mean, what we’re seeing, really, is, as CJ said, in the financial sector, we’re seeing the application development really accelerate around blockchain technologies today, and what these organizations are doing is using it as an opportunity to make the changes in their organization around digital transformation. They’re doing it in the cloud, they’re doing it using open toolchains in the cloud, and, obviously, the ones that I’m most familiar with are doing it using IBM’s Bluemix, IBM’s cloud, and our continuous delivery offering with open toolchains. And that means that they’re using a set of modern tools, like Git and other tools in the cloud, to do this rapidly, and they’re iterating very quickly.

The thing that we’re hearing from our customers is that they’re making changes in production daily, that they’re getting to their customers with changes in a matter of days after the customer makes these requests. And that’s really the big difference that we’re seeing, is how quickly they can meet their customers’ needs in this space, using templates to quickly get their new developers up, using an open toolchain so that they’re using the common industry-leading tools in each of the categories around collaboration, around tests, around source code management, around issue tracking, so that they can deliver that, using those common tools, using the technologies of blockchain to deliver to the customer quickly.

Shimel: Got it. And so, Cliff, I imagine that’s a big part of your focus and, CJ, you as well, in architecting, is making sure that Bluemix and the IBM cloud—it’s still called “IBM cloud,” right, Cliff?

Utstein: It is. It’s—I mean, we call it “the IBM cloud” and it means—you know, there’re a lot of elements of the IBM cloud, but “IBM cloud” is a good moniker.

Shimel: Okay. Just wanna make sure. So, you know, part of your job, obviously, and CJ as architect, is you’ve gotta make sure these underlying tools are there in place for people, as they onboard, as they start making this migration, this transformation. And many of them are, at the same time, adopting DevOps type of customs and culture and tools to facilitate this as well. So it’s almost like you have to stay one step ahead of anticipating what these guys are gonna need.

Paul: Yes. And that’s what—now we look ahead. We look at what some of the early adopters of these technologies have observed and we’re trying to remove the roadblocks, right, both in terms of providing sample applications, quick ways to get them started with their own enterprise applications, and then, as they get started building these applications—nobody builds an enterprise application as a lone developer; they always work in a team. So we wanna make sure that we enable these teams to be productive very quickly.

And that’s where, as Cliff mentioned, we support a open-toolchain fabric, unique in the industry, and then, on this fabric, we template-ize the toolchain that these things need to get going quickly, so we can instantiate those templates within minutes and, basically, onboard a team, with both starter code for their applications, as well as the toolchain they need to get it out into production, so we’re pretty excited about that. You’ll see those kinds of capabilities show up in the way you build these Watson applications and the financial applications and the IoT applications on Bluemix and on IBM cloud.

In fact, what we tell customers is “DevOps is the way you build any application on the cloud, given that all of the infrastructure can be provisioned through APIs—you know, it’s all infrastructure, this code, on the cloud—so you don’t have to wait for your contracts to close. You can just make API code and get your infrastructure and provision your runtimes and deploy your application. So there is no other way to do it but to use DevOps.” And that’s why we are working hard to enable all of these developers to have an outstanding experience on IBM cloud, working the entire developer journey, from how they get started with the sample apps, how they get started with their enterprise apps, and how they build and deploy with toolchains.

Shimel: So let me just repeat that one line ’cause that was the money line, CJ: “There’s no other way to do it but to use DevOps.” I love it. Cliff, what do you think about this? Are you in violent agreement with CJ? Anything to add there?

Utstein: I’m in violent—oh, CJ and I are always in violent agreement, Alan. Yeah, we’re seeing—and I think, as these companies move to the cloud, as they take advantage of Watson and blockchain and IoT, it’s critical that they adopt DevOps practices or else they’re unlikely to succeed. It’s gonna take them much longer to get to market, they’re gonna do it with less customer satisfaction, lower quality, and they know it. I mean, the data’s in. I read it on your website, Alan. Go to devopsy.kinsta.cloud; you’ll see.

Shimel: Absolutely. Guys, we’re kind of running a little bit low on time. We have a few minutes left. I wanted to talk—we had IoT on our list and I don’t wanna shortchange you because it’s important. You know, IoT, to me, guys, is—I mean, there was so much hype around this, let’s say, a year or even two years ago. We went suddenly from IP4 to IP6 because every refrigerator’s gonna be connected, but it’s not just the refrigerators and toasters, right? It’s all of the little gadgets and how many IoT devices are on our cars, even if we don’t own a Tesla. How many—you know, I signed up for a contest yesterday to get one of those free RFID tags that you can put on your keychain so you won’t lose your keychain anymore, and it’ll show you where it is or your wallet or anything else you wanna find. And, to me, you know, even in my house, we’ve gone smart home and I’ve got all these IoT devices now that I can control from my phone. So the hype is real, but, first of all, how much IoT are you guys seeing with the cloud and in the IBM customer base? Either one of you.

Paul: Yeah. I would say the IoT scenarios—you mentioned smart home as an example, but connected cars are another big driver because now you can buy cars that have built-in WiFi plans, right, that are paid for, for some number of years, by the manufacturer itself. So you can see maps and understand, for example, what the traffic durations are or there’s burglars in an area, even as you do navigation on your car navigation system.

So lot of these connected devices both pull and push data to the cloud, and that is where a lot of the data collection and analysis applications are actually being built out, to support these devices on the edge, because lot of the compute that backs them up is out in the cloud, right? And so, again, all of these are brand new applications that nobody had before, so they have to build it and iterate on it rapidly and get it out and get feedback from the users and continue to improve it. And there, again, all of the practices and the toolchains are very relevant to any of the teams building the IoT applications.

Shimel: Sure. Cliff, what do you see on your end?

Utstein: Yeah, I mean, as CJ is saying, I mean, automative is a big driver right now of IoT and BMW’s CarData platform was created to run on the IBM Bluemix cloud, in fact.

Shimel: Oh, right.

Utstein: That was just announced two weeks ago. And it really gives the new platform access to IBM Watson IoT capabilities, so that’s a big deal in the space and they’re talking about 8.5 million BMW customers globally taking advantage of this service, and it’s really about data. I mean, as we talk about IoT, it really comes down to big data and needing a cloud to manage the big data and needing practices like DevOps, if we tie it back to DevOps here. It’s really about practices and giving developers the ability to take advantage of these practices to build that relationship with the customer so that the customer can make sense of that data ’cause that’s what the cloud can do best.

Shimel: Yeah. I mean, my sense, guys, is that, with all—you know, CJ, you mentioned a few applications. Cliff, you literally put the rubber on the road with BMW relationship—you’re right. But I still feel like we’ve barely scratched the surface between how IoT—connected devices, if we will, interacting with the cloud, are gonna change the fabric of our lives. Right? And I think we haven’t even touched on the medical aspects of it, when you have nanotech running in your body, reporting back somewhere to the cloud about your vitals and your bodily functions and so forth. I mean, it’s gonna change things.

But, Cliff, you hit on something that I think is really important about all of this. Underlying all of these things is big data and big data analysis and analytics, and that’s where cognitive becomes the game-changer, in my mind, anyway. I mean, what do you think, Cliff?

Utstein: Yeah, I think that’s it. I mean, I think IBM’s cloud bringing together the platform for big data and cognitive together, with a set of DevOps processes, I mean, is truly differentiated.

Shimel: Absolutely. Guys, you know, I promised you 15 minutes; here we are at 22. So we’re gonna need to wrap things up here, but, you know, each one of these areas—cognitive, blockchain, IoT—probably, we could have done a podcast on each of ’em, but I’m glad we were able to hit what we did. Maybe we could come back on in another time and part-two this and dive a little deeper ’cause I did wanna touch on the connection between, let’s say, cognitive and IoT, and I think we’ve just scratched the surface. But, for now, we’ll call it a wrap on this edition of DevOps Chat. Cliff, thanks so much. I know it was your first time on; I hope we didn’t scare you away and you’ll come back and join us again.

Utstein: Thanks, Alan.

Shimel: Thank you. CJ –

Utstein: Would love to.

Paul: Thank you, Alan.

Shimel: CJ, always a pleasure having you on, my friend. I look forward to seeing you soon.

Paul: Thank you, Alan. It was a pleasure.

Shimel: All right. CJ Paul, Cliff Utstein, IBM’s cloud and Bluemix team. This is Alan Shimel for devopsy.kinsta.cloud and DevOps Chat. Thanks for joining us and we’ll see you soon—or listen to you soon, on another DevOps Chat.