I recently had a chance to sit down with Jeff Schaeffer, senior vice president and general manager of Continuous Delivery and DevOps Business Unit at CA Technologies. I have interviewed Jeff before, most recently at Jenkins World, where we did a video on our DevOpsTV YouTube channel. Jeff is a great interview who always gives good insight into what CA’s and his views of the dynamics shaping the DevOps market are.
Jeff and I had a chance to talk about the launch of the new CD neighborhood sponsored by CA Technologies. Specifically, we discussed how CD—and, for that matter, the whole idea of continuous IT—is reshaping the way organizations do business. It is a great conversation that I hope you will enjoy.
Below is a streaming audio of our DevOps Chat and below that is the transcript of our conversation.
Alan Shimel: Hi, everyone. This is Alan Shimel, editor in chief, DevOps.com, here for another DevOps Chat. Today’s DevOps Chat guest is Jeff Schaeffer, who is responsible for CA Technologies’ continuous delivery CD business.
Jeff, welcome to DevOps Chat.
Schaeffer: Thank you, Alan. Glad to be here today.
Shimel: Good to have you here. Jeff, of course, I think the last time you and I spoke was at—it was at the Jenkins World event, and we had a chance to actually sit down on video and talk a little bit about, you know, where do you see the market, where you see the DevOps market going, what CA’s involvements are and where you guys are kind of placing your bets and moving your pieces around as this market continues to evolve.
And we were, of course, talking offline before we started today’s podcast about kind of how you view the different waves, if you will, of maturation of companies’ life cycles with software and so forth, and you mentioned third wave, fourth wave.
I’d love, if you don’t mind, if we could share a little bit of that context with our audience in terms of your view of the historical ebb and flow here of history, if you will, of where we are. So would you mind sharing that?
Schaeffer: Thanks, Alan. So, Alan, as we were discussing, our—the thesis and the drivers for continuous delivery actually start kind of in the history of technology, and it’s probably never been a more exciting time for any of us to actually be involved in application development and technology in general as it’s transforming all the markets.
And the history is useful to the extent that it helps us understand the future as well, and when we think about it, you know, we’re really in the midst of the third wave, in which businesses have disruptively adopted technology in the business.
The first wave was back in the 1960s and it was really about the mainframes and the digitization of the paper processes in the company. I mean, it was a process disruption.
The second wave was in the ‘90s, and it was really about ERP and CRM systems where we were automating the supply chains of companies, and again, it was a process disruption.
Today we’re in the third wave, around the application economy, and in the context of digital transformation and the app company, you know, companies today, they promote their brands, their products, their services through their web and their mobile applications, and it’s become the primary methodology by which they engage their end users. And it is actually underpinning and enabling business model disruption, which is very different than the prior two waves of disruption.
So when we look at it, it’s an entirely new business model. We’re all familiar with the majority of those in terms of, for example, Uber disrupting transportation and Netflix disrupting media, et cetera.
And one of the challenges about the third wave is that—following the third wave, actually, I think is the fourth wave, and the fourth wave is gonna be around systems of insight and engagement, by which companies are able to take all of the multitude of data that they’re generating in terms of every interaction with their end consumers—you know, with suppliers and providers—and be able to analyze those data sets’ multivariate analysis and derive deep insights, and to be able to then, in turn, customize their offerings. Companies would be able to use that data to make more of the right decisions more often, to drive efficiencies, et cetera.
And so naturally, the fourth wave—and it’s coming, but before we can actually fully benefit from the fourth wave, we actually still need to master the third wave. And in the context of the third wave, the app economy, it’s really about, how do you get these applications, these web or mobile applications into the hands of end consumers more quickly and, you know, great quality and at lower cost and factor of production.
And so that’s really the area where companies today are focused. And, to do that, what most companies are doing is, they’re adopting agile methodologies, right? They have to increase their competitiveness and their ability to deliver this custom software. So they look to get development up to velocity, so they’re adopting agile methodologies, and then they quickly realized that the rest of the value streams are not at velocity—so testing, release, operations—which then leads them to adopt DevOps, which is where we’re gonna have a great conversation today.
And, with DevOps, they start to adopt the culture and the practices, and then realize they need a new enabling tool chain, and so if you think about any technological disruption and being a function of organization and people being a function of process and being a function of tools and technology, they now adopted a new tool chain, which is the continuous delivery tool chain.
And that’s an area where we at CA are focused about how do we provide the coaching around the organizational transformation and the practices in the process transformation and then, you know, from a product perspective, we’re building the next generation tool chains, building block fashions, and bringing those to market to support companies’ transformation and competitiveness in the application economy.
Shimel: Perfect. And you know what, Jeff, that actually dovetails into, of course, just today, we announced on DevOps.com that we’ve added a new CD—or continuous delivery—Neighborhood, which CA has graciously sponsored. And we’re really looking to build this into a—I don’t want to use the word portal, it’s so 1999—but a destination where people can come and get some great content and some great education around this whole CD tool chain and the CD economy, if you will, that is growing up.
But, you know, Jeff, going back to what you were talking about, we can’t get to the fourth wave until, I think, the third wave is a little bit more solidified and a little bit more mature, a little bit more settled. In my mind, what we’re seeing is, a lot of these convergent circles starting to overlap, right? You mentioned agile, we mentioned CD, we mentioned things like API, right, which is another big area for CA.
Schaeffer: Yes.
Shimel: And, you know, they’re not totally independent of each other, nor are they totally dependent on each other, but you know, they kind of form the ultimate Venn diagram where the overlap is just marvelous, right? I mean, it’s just a wondrous time to be in business if you’re using—you know, you’re leveraging all of these things.
You know, and you’ve been around, Jeff, as I have—what do you see with that?
Schaeffer: Yeah, so Alan—exciting things, exciting things. So, a couple of quick comments, one on the Continuous Delivery Neighborhood—really excited for that. Alan, I can tell you that in that context, whenever I meet with peers across the industry, one of the things that comes out very broadly today is, there’s a big gap between conceptualization of things like agile, DevOps, and continuous delivery, and translating those into real world practice.
Shimel: Mm-hmm.
Schaeffer: And the opportunity to provide that education, to provide a journey to share examples of how people have adopted these and succeeded is gonna become absolutely instrumental in terms of unlocking the full potential of these methodologies and these concepts and deriving the business benefit.
So we’re really excited with what DevOps.com is doing with respect to the educational exercise, and also DevOps.com and that sponsorship of the DevOps Express Consortium, which is another multi-vendor body that’s focused on this translation of concept of real world practice, and of course, here at CA, we’re happy to be a founding member of DevOps Express as well as a sponsoring member here for DevOps.com and the CD Neighborhood.
The second thing is, when we look at the change in the continuous delivery tool chains, what we’re finding is, as companies adopt agile, they start looking to have—okay, what is the prescription of the next generation software factor, software life cycle, and as they look at continuous integration to continuous test, continuous release and deployment and continuous operations, you know, how do you string these building blocks together?
And the only way to reduce the cycle times is making both a process change but also the tooling change. And the tooling is very different today versus, say, some of the legacy providers of application development and testing tools. It’s all pragmatic, API driven, so that it’s fully automatable. There’s a lot more—as you were describing the intersection and the Venn diagrams, it’s really a function of the feedback loops. So how do I take data, for example, from operations and use that to make a smarter example in continuous testing?
Let me share an example of what we mean by that. So, if you have a mobile application, you know, mobile application is much more difficult to manage in production than, say, a web app. First, people simply don’t update their smartphones that often, and so you end up with a long tail of versions in production environments.
Shimel: Mm-hmm.
Schaeffer: They’re running on different operating systems—iOS or Android and different versions of the operating systems. They’re running on different hardware platforms like an iPhone 4 or an iPhone 7, Galaxy 7. They’re running across different carriers, and there’s over 700 different carriers worldwide. And it becomes a factoral problem, and it’s unlikely that any given testing team has the ability to model that many combinations—because it’s a factoral problem—to model that many combinations in their test matrix.
But, if you have operating data, and you know that 80 percent of the user base is on these five to seven combinations, then you can provide that data back to the QA team and proactively test those combinations and take your risk in the tail of where people have more infrequently completed combos. But, by the same token, if you can detect failures in operations, you can provide that same data back to dev, and they can focus their investigative research as they work on the current product engineering.
So when we think about the Venn diagrams and how all of that works together in the future, it’s—not only are the format factors of the tools changing, but the feedback loops become absolutely essential in order to fully automate and to systematically improve the software life cycle process. And so, a really exciting time.
And we’ll often hear that today, this idea of shift left and shift right, but it’s really a—you know, those things embody, that vocabulary embodies a number of concepts. One is, you know, how do I push activities earlier and earlier into the development process under shift left, so how do I incorporate, say, for example, performance testing early, and then how do I tie performance testing into operational testing, and how do I provide that feedback loop all the way back?
So just exciting times, lots of transformational changes, and people are definitely looking for prescription and methods as how to reconfigure their software life cycles and incorporate that into their current software development methodologies, and that’s what, I think, DevOpsExpress.com will be so good at is helping provide that center of knowledge and best practices for the betterment of the industry.
Shimel: Absolutely. Jeff, believe it or not, we’re almost up at time, but I wanted to—and I don’t mean to embarrass you, but I want you not to be shy here. CA has done a masterful job of amassing, developing—both holistically, acquisitions, et cetera—a play in this new third wave, fourth wave economy, if you will.
Schaeffer: Absolutely.
Shimel: Give our audience—again, don’t be shy. Why do you think CA is best situated to really leverage what’s happening out there right now?
Schaeffer: Wow, great question, Alan. So, two, maybe three key points on that. So number one, Clayton Christensen, who is a professor at the Harvard Business School, he’s written extensively on the concept of disruption and then disruptive innovation in the market. And one of the key things that you learn about, or learn from Clayton, is that the incumbents never survive the disruption. And it’s not because they’re—you know, it’s just because of their decision making and how they’re driven by their paradigm.
And CA, uniquely, is in a position as a company to invest in this next generation of continuous delivery capabilities. And so we’ve been putting together this next gen tool chain through a combination of organic product development as well as acquisition, and so we’re uniquely in that position in the market in terms of being a vendor of sufficient size to provide that consolidated next generation tool chain and to ensure it works together. Per kinda the research of the market and industries over time, you know, we’re in a unique position relative to the traditional incumbents and I’m excited.
You know, we’ve pulled together the Rally Software, which brings kind of market leading agile management capabilities into the portfolio and we start with the planning process and agile management. We realize that, in the application economy driving the automation, that oftentimes in the app economy, the API is the application, but that’s also the primary means by which we exchange data and are able to integrate or automate all of the functions as well as create the next generation architectures, which are all kind of composite applications built on microservices which work through APIs.
So we’re connected by the APIs, and so we’ve invested in their seven and brought in a really robust API management capability, and then we are laser focused in the continuous delivery space. So, how do we take and orchestrate release with release automation software and release automation continuous delivery addition, which is basically harvesting the data and providing the analytics so that we can systematically improve and continuously improve the software development life cycle.
And then, beyond that, we have a number of kind of the next generation continuous testing tools. So, everything in terms of, we just recently acquired BlazeMeter. It’s a fantastic capability and at the heart of continuous testing. You know, it’s a commercialization of open source projects, JMeter and Taurus, which allows us to automate any open source testing tool, whether it’s the JMeter testing, which is the second most widely adopted community developed testing product, or Selenium, which is the most widely adopted testing product. Gatling, Locust, et cetera, so a host of open source testing engines.
We add value above and beyond that. It’s all hosted in the cloud and deploys both in public and private clouds for the tests themselves, and so it’s highly automated, it’s very scalable, and it allows us to left shift the testing all the way into the integrated testing, the component testing, and the unit testing of the new life cycle processes.
We also have put together a number of tools that improve testing efficiency around test data management, so how you can’t automate your tests unless you automate the test data, and providing test data with the right sort of attributes in order to drive the tests in a productive way. We have service virtualization software, which allows us to simulate test environments, and to emulate the test environments and abstract away things like mainframes or databases, which allows for both parallelized development. When other components aren’t available, we can simulate those or emulate them, and then in the testing practice itself, we’re able to isolate applications and put them under load testing in an effective way.
All the way out into our operational tools, where we’re able to shift right, and provide kind of the—you know, what’s done in pre-prod gets used in production, how do we get the integration and the feedback loops back into the testing and the development processes.
And so we’ve really been focused on putting together that next gen toolset around continuous integration, continuous testing, continuous release and deployment, and continuous operations. And again, we’re in a unique position to do it. We’re unique in the fact that it’s a disruption within the market and we’re able to put together this tool chain, and we’ve made a lot of progress in developing that tool chain up ‘til now. And we’re finding companies like—the people we’re working with across the industry adopting the CA tool chains, you know, we’re finding some significant improvements in not only the velocity but the quality and the cost effectiveness of their software development practices.
So, for example, companies that were getting one release out every year are now able to do 30,000 deployments per month on 300 applications in both the test, the pre-prod and the prod environments. We’re able to significantly reduce the cost of handling data and increase the automation of testing, up international the 85, 95 percent range, lowering the costs and achieving ROIs, in some cases, of over 300 percent in paybacks on the investment measured in months—you know, two to three months, which is very unusual around what people would normally expect in an enterprise software vendor.
So we’re really excited with the difference we’re making in the lives of our customers—the operating benefits, the business benefits, and then the overall, you know, the impact that we’re gonna make in the industry as a whole.
Shimel: Fantastic. Hey, Jeff, we’re way over time. I apologize, here. I apologize to—well, I’m not apologizing to those listening, I think they’ll find it worthwhile. But we do try to keep these at 10 to 15 minutes, and we’ve gone over—but I think it was worthwhile, we had a lot of information to share.
Jeff, continued success in leading up the CD business at CA Technologies, and we’re looking forward to more exciting news soon, and we’re actually looking forward to the CD Neighborhood on DevOps.com really becoming a go-to place for people looking for information around it. So thanks very much on the sponsorship for that as well.
We’ll be in touch again, soon. Until next time, I hope you stay well. Jeff Schaeffer, he runs the CD business at CA Technologies—thanks for being our guest on this episode of DevOps Chat.
Schaeffer: Thank you, Alan.
Shimel: Thank you. This is Alan Shimel for DevOps.com, and we’ll see you again on another DevOps Chat.
