Glenn Sullivan, Infoblox | Next Level Network Experience
(relaxing electronic music) >> Narrator: From around the globe, it's theCUBE! With digital coverage of Next Level Network Experience event. Brought to you by Infoblox. >> Welcome back to theCUBE's coverage, we're here in our Palo Alto studios. I'm John Furrier, host of theCUBE. We're here with Infoblox for their Next Level Networking virtual event with theCUBE. Glenn Sullivan is our guest, Principal Product Manager with Infoblox, formerly with SnapRoute, theCUBE alumni. Great to have you back on, Glenn. Great to see you, and thanks for jumping on remotely. We're doing the remote thing, the remote CUBE, good to see you. >> Yeah, it's great! I wish I could be in the studio, you guys have a great studio up there in Palo Alto, so I wish I could have joined you, but that's not possible right now. (chuckles) >> The governor's on, is off, we're get there, but when it does come back we'll certainly do a lot more remotes, and want to go to a "hybrid world." Hybrid, it sounds like the media business is turning into cloud computing, you got public videos, in person, you got hybrid, and virtual. The cloud native world is certainly spawning everywhere now with COVID, and you guys are talk about Next Level Networking, but with the word Experience. I want to get your thoughts on that because, you know, it's been six months, you've been on theCUBE, a lot's happened. Next Level Networking Experience, describe it. >> Yeah, it's really about processing things as close to where they need to be processed as possible, right? So, you don't really want to put everything in the cloud, you don't really want to have everything happen on-prem, you want to do the right data processing where it's needed, right? Have a little bit on-prem and have a lot in the cloud, or vice versa, it's really about elastic scale, right? That's what I think about with cloud native technologies is being able to run whatever you need to run service-wise as close to the delivery mechanism of either the user, or you know, as close to the app in the cloud as you need to. That's really what it means by, you know, having an elastic scale, and we try to do that every day. >> And notice the word Experience is in there, you know, that's been super important because you build and provision, manage these services from the customer standpoint. I mean, I can't drive in, there's no, there's clothes, or I got to go in, I now can do remotely. This is the key about having abstraction layer innovation, certainly DNS, DHCP, IP address management, never going away, you've got to connect stuff to the internet, I mean the network is there. >> Exactly. >> You've got to be a bit more innovative, what's your thoughts on the impact of the network now that cloud native and open source specifically are driving more action. >> Well, there's a lot going under the hood, right? And you can't just, you know, manage things the way you used to be able to, where you take and you buy a box, you know, it's that cattle vs pets thing that we talk about in cloud native, right? Where you treat this appliance very specifically and very specially, and you upgrade it and you're afraid to touch it. Now that you can't, you know, get the things, you have to do everything lights out. So, what we've learned via applying technologies in the cloud, you know, you didn't go into AWS' data center, or Google's data center, or Microsoft Azure's data center and manage these things, so what we've learned about how to manage infrastructure across the board in networking and compute and storage now is even more important, because everybody's lights out all the time now. >> And scale and speed is critical. I mean, Google's pioneered the concept of SRE, Site Reliability Engineer. What your teasing out, Glenn, is the same kind of concept for the network, you've got to have the security, you've got to have the scale. This is a huge point, can you react to that? >> Yeah, it's about spinning up instances where you need them, you know, when you need them, right? If networking equals a physical black box appliance that you specifically nurture and manage instead of just networking services, right, because DHCP is a networking service, DNS is a networking service, IPAM is a networking service, so you should be able to spin those up wherever you need to and manage those without having to worry about it all being tied to, you know, specific things that you have to manage in a very nurtured way. >> I want to get your thoughts, the term borderless enterprise is being kicked around, you guys use that term. I've heard, you know, the borderless networK, makes sense I guess, but what does the borderless enterprise mean to you. >> Well, it's really just an extension if you think about it from the software defined perimeter concept before. You know, people call it different terms now, but it's just saying that borderless means that I don't have people sitting in a office anymore, and if I do have people sitting in an office, they have the similar experience to people that are connecting remotely, no matter where they are. So, because there is no boundary to your network, right, because the edges of your network don't match edges of your walls in your branches, that's pretty borderless to me, right? And you have to kind of think about, you know, it's not just about adding more firewalls, It's not just about adding more network perimeter security, it's really about how do I apply foundational security across the board. I've been at Infoblocks now for a little over six months, and I can tell you, it's great to see thinking about these foundational services, right? These infrastructure services like DHCP, DNS, and IPAM being really at the foundational layer of the security that you apply to your network. Right, it's the first couple of things that happen, right? The first thing you do is you get an IP address, that's DHCP, you can figure out all kinds of stuff about a device that way. Then you start looking at services with DNS, right? And then it's like, "Okay, well now I've got a lot more information about what the user's doing, where they're going, and how to secure it," right?" So, these sound like they're really your plain vanilla protocol suites, until you really start applying borderless security across the board with them. >> Yeah, a lot of machinations, and also you now have massive amounts of connection points, 'cause with IoT, not only have more in terms of volume of things connecting, but they're being turned on and off very quickly. They have to get connected, so you have that going on. >> Yep, and then you got to make sure that they do what they're supposed to do, right? If they're supposed to phone home to a specific place that they only do that, and that they haven't been hijacked, and somebody isn't mimicking them with malware. There's all kinds of security threats when you start thinking about all the possibilities that IoT brings into account. >> Yeah, some light bulb that you screw in, wifi enabled, has a multi-threaded capability, and be, who knows what's on there, right? (laughs) I mean this is what the reality is, no one knows what connects, a little hygiene comes a long way. I want to just get back into what you said. You've been there for a few months, came from SnapRoute, which was doing some real fine work, that's where we did our feature interview on you and what you were doing there, that technology. With borderless enterprise, what is the role that cloud native and open source play? Because this is your wheelhouse, I want to get your thoughts because when you had that to borderless, things kind of happen. >> There's two things that I like to think about. One, it's scaling things down as skinny as possible, or as big as necessary, right, elastic scale, right? We talk about cloud native technologies, we always talk about elastic scale. Well, what does that mean? Well, that means that am I securing an entire data center? Am I securing a branch office? Am I securing a gas station? Or am I securing a person working from home? You know, this is what we mean by elastic scale. It doesn't mean that I'm, you know, purpose building the spoke specific security profiles for those individual use cases, it means that I have a system that I can scale up and scale down no matter where those folks are, right? That's really what you have to do when you think about cloud native technologies and the borderless network, is you have to be able to run things as close to the user as possible, or as close to the app as possible, or somewhere in between. The second thing that I think is super key is abstraction, right? You can't manage everyone working from home, or you can't manage as many instances as you need with everyone's individual laptop, right? This doesn't scale, right? Abstraction is key to cloud native technologies because it means that I don't pay attention to anything that's below me, right? If I'm an SRE, I don't necessarily care about what type of servers that application set's running on. If I'm a network engineer, I don't really care about the fiber patch panels that connect my network devices together, right? Abstracting away the underlying infrastructure is key for cloud native technologies. So, as we add more and more devices, more and more endpoints, more and more users to manage, we have to make sure that we abstract away the complexity of all the connections that need to be built between those users and whatever, you know, abstraction orchestration layer that we utilize. >> You almost peeled back the onion from the early days of DNS and go to the core, "Hey, I want to connect to this domain." And a packet moves from here to there across an IP address, "Oh, let's add some abstraction on it." This has been the innovation form for the internet for years, right? So, how do you describe the Next Level? Because you mentioned, again, the word Experience is in there, so Next Level means, okay, networks need to be programmable. You do have the Next Level opensource dynamic that you pointed out beautifully, what's that Next Level Experience? How do you see the preferred future evolving? Because if you take this further, if you believe cloud native provides some scale, as you pointed out, it should simplify, these abstraction layers should reduce complexity, or abstract away the complexities and provide more simplicity. >> Absolutely! I mean, I always come at it from an Ops perspective because that's just my background, right? But I was running networks for a long time before I started building, you know, network operating systems, right? I can tell you that what I need is visibility. You know, I need to be able to see what's going on at any given moment. I need to be able to know that the things that I've deployed are up and running. I need to know that the information that I need to troubleshoot the issues that arise is at my fingertips, right? Because I always think about it like the 3:00 a.m. call, right? The network engineer, or sysadmin, or the DNS admin, or it doesn't matter who they are, at 3:00 a.m. they got to wake up because they've just been paged, and something's wrong. And how do they get to what's broken? So, that's one way to think about it. There's also the deployment way to think about it, right? Like how can I deploy as many new users, as many new branches, as many new locations, whatever the process is. You know, you hear zero touch provisioning, you know, all these other, these features, and they come as part of a cloud native mentality, right? They mean that I don't have to do, you know, a whole lot of pre-thinking and pre-staging, and pre-configuration, and pre-thought before I deploy stuff, right? It means I need something, I deploy whatever is required from a service level, I kickstart it, it bootstraps itself, and it joins, right? I take away the headache of having to think about where something is or when it is, and that's a lot of the synergy that we had between what we were doing at SnapRoute and when we came to Infoblox, right? I can tell you, we were pleasantly surprised by the platform that was built, and we were like, "Okay, well this is going to be great! We can add services to this and we don't have to worry about having to go an reinvent the wheel." Because when you choose technologies like Docker containerization, you choose technologies like Kubernetes orchestration and Kubernetes abstraction, you are a lot closer to where you need to be. I mean one of the thing that, you know, isn't super well-known out there is that CoreDNS is one of the major projects that Infoblox helps maintain with inside CNCF, the Cloud Native Compute Foundation, right? CoreDNS ships at the core of every Kubernetes version from now on, you know, as of a few versions ago. So if you think about it, Infoblox has got a lot of cloud native technologies built into everything that we do, and we're one of the key maintainers of one of the key DNS features of something that's at the heart of Kubernetes, and you know, I don't have to tell you how popular Kubernetes is. >> Yeah, we've chatted about that. It sounds like it's the kernel of all the action, DNS, the CoreDNS for Kubernetes. (laughs) >> Exactly, exactly! It's definitely at the core there. >> Glenn, I want to get your thoughts. First of all, I love chatting with you, you mentioned you were from an operating background, but also you can bring a lot of dev into it too, so this is ultimately, to me, the inflection point of where DevOps goes mainstream, because you used to do Ops for a fruit company, Apple? >> Yes, yes, very popular! >> Big one. >> A very popular fruit company called Apple, and we know how hardcore they are, especially they lean heavy on, you know, lock it down, make sure everything's secure, I mean it's well known in the Silicon Valley and around the world, certainly in tech circles, the security mindset. >> Absolutely. >> Large scale operations. Now, you bring also the DevOps aspect of it with cloud native. As that world has to become secure, and networks, it's an Ops game, let's face it. No matter how much DevOps you sprinkle into the equation, at the end of the day, it's Ops. Ops, operations of networks, high availability, large scale. But now you have a little bit of development goin' on on top. The programmable internet past the tip of the network layer, what's your take on that? Because you still need security, you want to have the capability to do some advanced automation. These have been hot new trends, and networking people are now hearing this not for the first time, but it's the new thing where it's like, "Okay, I can have my Ops, but I got to do some Dev now." So make sense of this, where are we in this whole programmable networking aspect? >> Yeah, there's sort of two schools of thought, and it's interesting what's happening, right? You've got kind of, on the extreme left side you've got, "I just treat the network like it's dumb plumbing and I run all of my software overlays on top of it, and I basically treat the network like it doesn't exist." And you know, it's kind of a situation that's been perpetuated by the silos that are out there, where you have the network engineers, and the server compute engineers, or SREs, and then you know, it's like, "Well, these folks never have to talk to each other because we just treat the network like it doesn't exist, and we run overlays on top." And some of the vendors in the server overlay security space have been really proud of that interaction. And I can tell you that that's one way of doing it, but it's not the optimal way, right? Like, when I was a network engineer I could tell you, you're trying to build credibility, right? So, if I was talkin' to a network engineer now, and I'd say like, "How do you get your credibility built with your server folks?" It's kind of like learning a different language, right? If you try, if you try to speak the other language, the person actually is appreciative of that and will help you. So, I always found, you know, find thing things you can automate, run that code base, figure out the API structures, build some pseudo-code together to make it happen, and figure out what you're doing over, and over, and over again and automate it. Automate away, right? And that's some of the nice things that are the same here, right, everything we could ever want to do in any GUI is all REST API'd underneath the hood, right? So it's like, we don't have to pitch to people that, "Oh, you can automate this code if you want to, you can run these APIs if you want to." They know it, and they use it, and people are happy with it. And I think if you're a network engineer, you've got to spend the extra effort to try to, you know. You don't have to do anything complicated! >> It's not rocket science. You know, it's not like you got to go right C, I'm sorry? >> It's not rocket science. >> No, start with Ansible, you'll learn some Python, you'll learn some Django on top of that, and then keep running, right? Keep automating on top of that. >> All right, great stuff, Glenn. I know you've got a a tight deadline, appreciate you comin' on for this virtual fireside chat as part of the Infoblox Next Level Networking virtual event. What specifically can companies do to get what they need from a technology standpoint to secure the borderless enterprise? How do you see it playing out, now that you're on Infoblox side from SnapRoute, with what Infoblox has, which is a holistic portfolio approach, a holistic view, what are you guys offering customers, and how do they secure their borderless enterprise? Really start with DDI, right? I know DDI is something that is not specific to Infoblox, but if you look at what we're doing with DNS, DHCP and IPAM, it's really the foundational layer to start securing the rest of your network. We don't necessarily make it so you don't need the rest of your security stacks that are running on top, but we do optimize 'em and we make it so you can right-size 'em, and we really think that if you focus on getting that layer solid, and you really focus on the DNS security, you can apply a lot of lightweight, high impact features as early on in the packet forwarding process as possible. Right, if you think about, I'm a network engineer at heart, so I always think about the path of a packet from the start to the end, and DDI happens really early in the process, so if you give that right, the rest of your security infrastructure built on top of that is just going to work that much better. >> You're the Principal Product Manager at Infoblox, formerly with SnapRoute, how do you fit into this? What product are you managing? Can you give a little bit of background, kind of what you're working on? >> So, I'm an emerging technologies PM, so basically anything kind of new and cool that we look to add to our platform, that'll come out of myself and my group. >> And Kubernetes obviously is one of 'em. >> Well, Kubernetes is already there, so we're already doing stuff with Kubernetes inside Infoblox, like, our whole platform. If you buy BloxOne DDI and BloxOne Threat Defense today, it's all deployed using Kubernetes and Docker containers, and orchestration layers, and everything today. So, everything that we're building on my team, is all building on top of that well sold platform that's already been developed. >> There's definitely demand out there, you're startin' to see the big companies like VMware, very operational focused companies start acquiring cloud native and open source, kind of a new kind of section to them. Obviously it's a tell sign, the markers are all there in terms of the trends. What are people missing? What's real, what's vape or what's reality when you look at the landscape, and what does Infoblox bring to the table? >> So, I think what's important to know is that when you're lookin' at open source technologies, a lot of them have been hardened over many years, and there's new stuff coming out all the time, and there's definitely new uses for them. But what's kind of important is what you put on top, right? Everyone's got open source under the hood, or they've got technologies they've OEM'd under the hood, right? But the experience that you present to customers is really key, right? Because you can take any kind of open source project and wrap a, you know, very thing layer on top of it, and you can either, you know, trump up the open source software, and say is the open source software we use underneath, or you can downplay it and say hey, this open source software, you know, we don't really talk about what's under the hood and it just all works magically. We find that transparency is really helpful. You know, you let people know what's under the hood, and you contribute to it, and you show that you're involved in this community, and you use that as a leverage to kind of push forward. So, if you look at, you know, what we're doin' with some of the different projects within, you know, BloxOne DDI uses Kea, and we're part of IC that's part of the maintainers of that, like we're openly in this space, right? And I already mentioned CoreDNS before, right? So, you can either take open source, and use it, and pretend that you don't, or you can take open source and contribute to it and be a community member, and be an advocate, and usually when you're on that side of the equation, you end up in a better place with your customers, building, you know, building confidence in your customer base. >> That's great stuff, Glenn Sullivan, thanks for comin' on, I really appreciate it. I'll give you the last word. In a nutshell, if I have cloud native and open source, how do I secure my borderless enterprise? >> Think about it as close to where the source is as possible and scale things elastically so that you can do as much processing of the user experience as possible so that you aren't trying to, you know, funnel everything to a single place and apply some magical policies in a single centralized location, to where you have to process a lot of data across the board. If you think about it from a hybrid approach where you've got a little bit on-prem and you've got a little bit in the cloud, or in some combination that's right for your organization, the hybrid approach that really trumps the local survivability, and really, you know, keeps focusing on securing things as close to the user possible, or as close to the source as possible, then you're going to be in good shape. >> Glenn, great stuff. As always, a masterclass in networking. Appreciate the insights, thanks for comin' on this Infoblox Next Level Networking virtual event for theCUBE. I'm John Furrier, your host. Stay with us, and thanks for watching. (relaxing electronic music)
SUMMARY :
Brought to you by Infoblox. Great to have you back on, Glenn. you guys have a great studio and you guys are talk about That's really what it means by, you know, you know, that's been super important the impact of the network and you upgrade it and can you react to that? that you specifically nurture and manage I've heard, you know, of the security that you and also you now have massive Yep, and then you got to make sure and what you were doing and whatever, you know, that you pointed out beautifully, I mean one of the thing that, you know, kernel of all the action, It's definitely at the core there. but also you can bring a especially they lean heavy on, you know, But now you have a and then you know, it's like, you got to go right C, and then keep running, right? and we make it so you can right-size 'em, that we look to add to our platform, If you buy BloxOne DDI and when you look at the landscape, and pretend that you don't, I'll give you the last word. to where you have to process a lot of data Appreciate the insights,
SENTIMENT ANALYSIS :
ENTITIES
Entity | Category | Confidence |
---|---|---|
John Furrier | PERSON | 0.99+ |
Apple | ORGANIZATION | 0.99+ |
Glenn Sullivan | PERSON | 0.99+ |
Infoblox | ORGANIZATION | 0.99+ |
ORGANIZATION | 0.99+ | |
Glenn | PERSON | 0.99+ |
Microsoft | ORGANIZATION | 0.99+ |
Palo Alto | LOCATION | 0.99+ |
Cloud Native Compute Foundation | ORGANIZATION | 0.99+ |
3:00 a.m. | DATE | 0.99+ |
AWS' | ORGANIZATION | 0.99+ |
SnapRoute | ORGANIZATION | 0.99+ |
Silicon Valley | LOCATION | 0.99+ |
Python | TITLE | 0.99+ |
two things | QUANTITY | 0.99+ |
Infoblocks | ORGANIZATION | 0.99+ |
theCUBE | ORGANIZATION | 0.99+ |
first time | QUANTITY | 0.99+ |
six months | QUANTITY | 0.99+ |
Django | TITLE | 0.99+ |
over six months | QUANTITY | 0.98+ |
second thing | QUANTITY | 0.98+ |
CUBE | ORGANIZATION | 0.98+ |
One | QUANTITY | 0.98+ |
Kubernetes | TITLE | 0.97+ |
first couple | QUANTITY | 0.97+ |
one | QUANTITY | 0.95+ |
one way | QUANTITY | 0.95+ |
CNCF | ORGANIZATION | 0.95+ |
First | QUANTITY | 0.94+ |
DevOps | TITLE | 0.94+ |
today | DATE | 0.93+ |
CoreDNS | TITLE | 0.93+ |
BloxOne | ORGANIZATION | 0.92+ |
two schools | QUANTITY | 0.92+ |
zero touch | QUANTITY | 0.91+ |
single place | QUANTITY | 0.9+ |
first thing | QUANTITY | 0.89+ |
Level | EVENT | 0.82+ |
single centralized | QUANTITY | 0.75+ |
COVID | ORGANIZATION | 0.75+ |
Threat Defense | TITLE | 0.71+ |
Next Level Networking | EVENT | 0.68+ |
Next Level | EVENT | 0.62+ |
VMware | ORGANIZATION | 0.62+ |
Kea | PERSON | 0.6+ |
Ansible | ORGANIZATION | 0.6+ |
Next Level | ORGANIZATION | 0.59+ |