F5 Networks has acquired NGINX, the leading open source web server player. The move clearly puts F5 in the DevOps space, as well as the open source market.
In this DevOps Chat, we speak with Lori MacVittie of F5 along with Sidney Rabsatt of NGINX. Some great insights into why this acquisition makes sense for both parties.
As usual, the streaming audio is immediately below, followed by the transcript of our conversation.
Transcript
Alan Shimel: Hey everyone, it’s Alan Shimel, devopsy.kinsta.cloud, DevOps Chat, and you’re listening to another chat. Today’s chat features two folks from an acquisition announced yesterday. In case you haven’t heard, F5 Networks has acquired NGINX. We’re going to hear more about it from two folks. Number one is our friend, Lori MacVittie from F5 Networks. Lori, welcome. Glad to have you here.
Lori MacVittie: Thanks, happy to be here.
Shimel: Fantastic. Joining Lori and I is Sidney Rabsatt. Sidney is VP of product at NGINX. Correct, Sidney?
Sidney Rabsatt: That’s right, and I’m happy to be here too.
Shimel: Well, everyone’s happy to be here. That’s not a bad thing. Are you happy to be here at the DevOps Chat or are we happy to be here talking about F5’s acquisition of NGINX? Let’s hope it’s both. Lori, give us a little background here. When I saw the deal yesterday. I said, “Wow, that’s a big deal.” I’m not sure – it’s certainly an expansion, up the stack, if you will, for F5.
MacVittie: Up the stack, deeper into the app. It’s all about application delivery. One of the – Sidney mentioned earlier when we were chatting, building on NGINX’s vision, which as I read it, is flawless application delivery. That’s certainly something that F5 has always aspired to achieve as well, but there’s just pieces in the application stack that we don’t play in, or haven’t played in until yesterday with the acquisition. NGINX is the leading webserver. They’ve got an app server. They do some app delivery. They’re very prevalent in many, many places. So we’re very excited to have them on board to see how we can bridge the gap between those two worlds.
Shimel: Absolutely. What about the perspective of NGINX? Money is always nice. Don’t get me wrong. What’s the attraction to F5 here?
Rabsatt: I think there’s at least two sides of it. Working with the folks on this early on, we got to get a sense of the culture of the organization. It felt very compatible. The people we met were great. We had a great time talking, hanging out, sharing ideas. There’s some real optimism about what can be done together. The thing that really gets folks excited, especially on my side of the house, on the NGINX side of the house is the opportunity to do even more faster. We’re a startup. It’s grown to a good size. The set of resources that we have, but looking at F5 and talking to those folks, understanding that they have a shared vision. They see the world very similarly and bringing more resources to bear to execute on the vision that we’re always working towards. It’s something that’s very exciting for us.
Shimel: I can’t disagree with you. I think we’re in violent agreement on it. It is a great opportunity. I guess the next question is how do we integrate, Lori or Sidney, how do we integrate the NGINX webserver, appserver, and the rest of the toolset into the F5 family? Or do we not?
Rabsatt: Do you mind if I take a stab at it?
MacVittie: Go ahead, Sidney, please.
Rabsatt: Sure. Look, I think on the surface of things it might seem as if there’s a ton of integration that needs to happen, and a lot of overlap needs to be addressed, but quite frankly, I’ve been pleased to see how much complementary capabilities we actually have. Oftentimes, in the marketplace, folks will actually deploy both solutions side by side. It’s because we do slightly different things, both super important to folks. When we think about integration, there’s opportunity for F5 customers to be able to leverage NGINX solutions in places where they couldn’t necessarily before. That should be a net benefit to them.
There’s the ability for folks that leverage NGINX solutions to take advantage of things that F5 does, that we don’t particularly do. I think that’s going to be another win for them as well. The integration is really focused on making sure we have the right extensions to the product line identified so we can really give customers the things they need to succeed in delivering those end to end applications, and yes, of course, we’re going to be looking to bring the best of both sets of technologies together in the places where it makes sense, but we’ve been pleased to see it’s not as many overlapping things as we might have found in some situations. We’re very optimistic about the path forward here.
Shimel: So crystallize for me, where do you see things that overlap versus things that play well together? Can you give us some specifics?
Rabsatt: Sure. Things that play well together fall into the category of “Hey, can I bring better scrutiny, better analytics, to the types of solutions that I might already have with NGINX?” On the NGINX side, I’m sorry – on the F5 side, “Hey, can I leverage solutions to give me a little bit more freedom and flexibility to build my apps wherever or however I might choose to?” So I think there’s some goodness there that folks will definitely get access to.
On the integration side, you’ll find most of it is just around making sure that the products work seamlessly together. So to the extent of I have multiple devices in my environment. How do I have a single solution that gives me the ability to manage it the way that I need to? We still need to go through the work of understanding exactly how the different buyers are going to need to interact with the different solutions that we use going forward, but all that work is in progress and again, the overlap tends to be the little stuff. Oh, do you draw a report? Oh yeah, we draw a report. Do you guys do that too? Sure. So how do we bring the best of those technologies together? It’s that kind of thing.
Shimel: Got it. Lori, interested to hear, you heard Sidney here. What do you add to that? What are your feelings on it?
MacVittie: I think he encompassed everything well. We already – you already see both of us in F5 and NGINX together in a lot of environments. From our perspective, I think we’re looking at – one, continuing to invest and amplify the NGINX vision and what they’re trying to execute on, but also where can we add security? Security is a big deal that has to extend from end to end. Right now, that’s a very disjoint process sometimes. That’s why we keep telling security to shift let. I think this is an opportunity that we’re going to be looking for to be able to do those kinds of things and shift that security even lefter with NGINX.
Shimel: Absolutely. It’s funny when you think about security being a driving force behind this acquisition. With NGINX being a webserver, that’s been an area and app server, that’s been an area where security has been required and important forever, even going back to IIS in the day, Internet Information Server and stuff. Here’s an area though I was thinking about. So many people in the public cloud use NGINX as their webserver of choice, whether we’re talking about Amazon or even Microsoft Google or any of them. Lori, does F5 maybe view this acquisition as a way to expand your footprint in public cloud? If so, how?
MacVittie: I don’t think we looked at exactly how that’s going to play out, and what we’re going to do with that. The webserver is something different from the proxy technology that most of our products are built on. So there’s some differences there in how that works. We’re still looking at roadmaps and trying to figure that out. For right now, we’re just going to keep going ahead with both sets of road maps and moving forward. As we continue to explore what we can do together, we’ll figure out where it makes sense.
Shimel: Sidney, same – flip side of that coin – of that question to you around public cloud. Do you see an angle where F5 helps there?
Rabsatt: Yeah, I think F5 already had strong relationships with quite a few of the cloud vendors as do we, and we expect that – we ultimately want to give folks a choice. This all boils down to organizations are trying to build the applications that make sense for them. They want to be able to leverage the best tools and capabilities that help them deliver those applications while giving them the freedom to build the application they want. So whether that is on one cloud, on multicloud, across clouds, and on prem locations, they need that flexibility to do what makes sense for them. A big part of what we’re trying to accomplish here is making sure that F5’s large customer base has the capabilities that NGINX large user base has, which is again, that freedom of flexibility. That means that whether it’s on Amazon Google, Microsoft, any cloud, or any on prem solution, they get to deliver their applications with the security, controls, the visibility that they desire.
Shimel: Excellent. I’m just – in hybrid situation, what about – one of the hottest topics certainly in technology today is the whole Kubernetes container thing. I know both F5 and NGINX had Kubernetes-related solutions and playing in that market. What, if any, does this acquisition, what if anything does this acquisition bring to that? Either one?
Rabsatt: I would say it should be a net benefit. NGINX, there was a recent report I saw. It’s in the public domain. NGINX is the most commonly used ingress controller for Kubernetes environments today. If you think about what Kubernetes means for organizations that are building these modern applications, it’s giving them an environment that allows them to build typically microservices-based applications. So getting those applications exposed to the outside world requires the first thing, the foundational element, the hierarchy of need. It’s getting traffic in and out of that Kubernetes environment. How do you do that? Through the ingress capability.
Our ingress controller is pretty much the number one solution out there for doing that today. That’s the first place people have to start before they get into some of the more complex or exotic things they’d like to be able to do. So our customers, our users are already benefitting from that. It’s allowing them to do what they need to do when it comes to adopting or testing Kubernetes for their own uses.
Shimel: Sure. Lori, even independent of NGINX, what effect is this whole move to Kubernetes and container-ized environment having on F5’s business?
MacVittie: Wow. On the business? Don’t really know. No. There is – we see a lot of customers asking. One, they want guidance on how to do this at scale and safely. A lot of our customers are very concerned about security, about performance. How do they bridge those two worlds, coming out of Kubernetes and trying to get into an environment where there’s a lot more traditional apps and infrastructure still supported? It’s trying to bridge those two worlds.
Our customers are looking at it and they’re bringing it in and they’re adopting it, but they’re also having challenges with the deployment side where NetOps traditionally sit that they don’t have the skills to do the automation. They have problems with integration of the toolsets to build those pipelines. So one of the things that this kind of an acquisition does is bridge that gap. Hopefully we’ll see better integration and easier in and out, both actually in the flow of traffic, but also just between the teams in the businesses, so we think this will help address some of the pains at the people level in organizations by having this end to end delivery chain with both F5 and NGINX.
Shimel: Sure. Sidney, maybe you would know. What are the plans with some sort of common interface or more integration between the two lines?
Rabsatt: I’d say it’s still very early on. We have some good ideas, but we’re not really in a position to preview too much. I will say there’s going to be a focus on again, making sure we’re bringing the best of both solutions together. Again, not nearly as much overlap as you might have expected, or we – I should say as we might have expected. Yeah, what we’re looking at is really the opportunity to bring more features to the things we already have, more so than trying to get rid of things. We’re looking forward to getting this innovation stuff underway. We’ve had some very early conversations, but over the next days and weeks, we’re going to be digging into it very deeply.
Shimel: Makes sense. Lori, so this was as we said right from the outset, what we’re turning to is a move up the application stack. What – is F5 done with that? Will we see more F5 potential either via the acquisition or holistically – organically moving up this stack here in the application world?
MacVittie: I can’t obviously comment on any kind of [laughter] –
Shimel: Can’t tell _____ acquisition is. I get that one.
MacVittie: That’s right. With any company, that’s always a possibility. We have been transforming ourselves in the past year really moving forward with a lot more innovation in terms of exploring things and looking at – we’ve got Aspen mesh going, which is service mesh trying to bring some sanity to Kubernetes inside for scaling. Cloud services, we’ve made a lot of our services as a service. You just hook them up. We’ve been moving deeper into the application stack for a while. This is just going to accelerate that transformation from ADC to application services company.
Shimel: Understood. Understood. Well, guys it really is exciting. One of the things I sit here at devopsy.kinsta.cloud and Security Boulevard and we go to all of the conferences and we cover all the news, and you hear so much “What’s bleeding edge, bleeding edge, bleeding edge?” We tend to forget, or we overlook what are the nuts and bolts, the bread and butter that keeps the internet and these applications going? It is things like the NGINX web server that is the dominant web server in the world today. It is things like the F5 lineup of software and of _____ actually that have been delivering the internet to the public since 2002 or whatever it is. Lori, you probably know, but so it’s good to see innovation continuing to happen. It’s good to see new combinations taking place.
I’m excited to see what comes out of – Sidney, as you said, it’s very early on. It’s a day after the acquisition was announced. Maybe we could check back in six months and see how things are progressing along these lines. In the meantime though, just the usual questions. Everyone from NGINX is staying on? There’s not going to be layoffs as part of the consolidations and stuff, or again too early to tell?
Rabsatt: I mean what we know is F5 is interested in our organization and not just products. They’ve made it very clear they want the vast majority of folks to stay and we expect that to continue to be the case. We look forward to what the future holds together.
Shimel: Yep, understood. Look, I wish you a lot of success, both you guys individually, as well as F5 and NGINX. It’s an exciting combination. Congratulations to both organizations and Lori, I think congratulations on especially to F5 for really making a real great addition to the family there.
MacVittie: Thank you. I know there’s a lot of us. We’re just really excited about this and about the future and what we can do together.
Shimel: Cool, cool, cool. All right, congratulations, guys. We’re going to call it a rap on this edition of DevOps Chat. Lori MacVittie for F5 networks. Sidney Rabsatt for NGINX, soon to be F5 Networks. Thanks for being our guest today on DevOps Chat. Thank you everyone for listening. You’ve just listened to another chat. Bye bye.
— Alan Shimel
Based on GitHub data, Adobe currently ranks as the 14th largest corporate open source contributor, up from No. 32 in 2017. One of the primary reasons Adobe was able to move up that list so quickly was its move to acquire Magento, an e-commerce platform based heavily on open source code. But now that the company has come to appreciate what can be accomplished by relying on open source code, organizations should expect to see Adobe accelerating the rate at which it will be making contributions to a wide variety of open source projects.
