Clicky

The Need for Observability in Application Development – Futurum Tech Podcast Interview Series
by Daniel Newman | August 21, 2020

On this special episode of the Futurum Tech Podcast – Interview Series, host Daniel Newman welcomes Karthik Rau, Vice President, Observability at Splunk, to discuss observability and application monitoring, COVID-19, and how IT departments are dealing with the recent changes brought about by the pandemic.

Karthik and I explored the rapid acceleration of digital initiatives due to COVID-19. Pre-pandemic, we were seeing companies slowly adopt digital transformation initiatives. But almost overnight, this progress was rapidly accelerated as companies were forced to go digital. We’ve seen a rapid adoption of SaaS solutions and other digital initiatives to support a remote workforce. It was a decade of digital transformation in three months. All of these changes have changed the way the IT department operates, and it’s most likely never going back.

The New Role of IT Departments. Karthik shared that as companies invest more in technology to support digital transformation initiatives, IT’s responsibility becomes more about enabling the organization to build the software that they need to drive their digital business forward as quickly as possible. It’s a significant shift to be more agile.

Rapid shifts in every industry. We discussed the impact COVID-19 has had on every industry. From education to healthcare to consumer goods, every organization had to pivot how they operate — most moving to a mobile-first environment. They’ve had to buy the technology and the software to support the changes as well as shift the culture to ensure the change.

Lastly Karthik shared about the impact of observability and the fundamental changes to app development. As companies have moved more towards SaaS, the underlying infrastructure has changed from on-prem to highly elastic public and private clouds. All of which is driving a larger DevOps approach creating massive amounts of data. It’s critical to avoid mistakes in the world of app development so companies need an observability solution that can track and monitor all the data that is created to catch errors and find trends.

If you’re interested in learning more about Splunk and how their offerings can help you, be sure to check out their website. Don’t forget to listen to the full episode below and while you’re at it, be sure to subscribe so you never miss an episode.

Disclaimer: The Futurum Tech Podcast is for information and entertainment purposes only. Over the course of this podcast, we may talk about companies that are publicly traded and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors and we do not ask that you treat us as such.

Related content from the Futurum team:

The Impact Of The Pandemic: Changes In Leadership Styles And Technology Approaches – Futurum Tech Podcast Interview Series 

Five Steps For A Data-Driven Future – Futurum Tech Podcast Interview Series

Splunk Posts Solid Results As Recurring Revenue Grows

Transcript:

Daniel Newman: Welcome to The Futurum Tech Podcast. I’m your host, Daniel Newman, Principal Analyst and Founding Partner at Futurum Research. Welcome to the Insider Series. Excited for today’s show. We have Karthik Rau. He’s the Vice President, Observability at Splunk, and he’s going to be joining me to talk a little bit about what’s going on with IT, the impact of the pandemic, observability, industries. We’re going to talk about all kinds of things, and it’s going to be a lot of fun. I’ll have him on the show with us in just a moment here.

But, as always, before I do jump into the show, I do want to remind everybody that this episode of the Futurum Tech Podcast Interview Series is for information and entertainment purposes only. While we do talk to executives from publicly traded companies and we talk about high tech and traded companies, this show is not intended to provide any financial advice. Anyhow, let’s get on with it. Karthik Rau, welcome to the Futurum Tech Podcast Interview Series. How are you doing today?

Karthik Rau: I’m doing great, Daniel. Thank you so much for having me on.

Daniel Newman: Yeah, I always wonder as people watch me do that intro, where they’re like, “When do I get to talk? Why is he still talking?” No, I’m just kidding. No, it’s really exciting to have you on. We’ve had some great shows. Splunk has been a regular here on the Interview Series. We appreciate the partnership. We’ve had your CTO, Tim Tully. We’ve had your CEO, Doug Merritt, and we’re excited to have you, who’s running the applications management at Splunk. Quickly, by way of introduction, since I’ve already told everyone your name and your job title, just tell a little bit about a day in the life at Splunk. What are you doing? What does leading application management at Splunk mean?

Karthik Rau: Yeah, sure. So I actually came to Splunk through the acquisition of the company that I founded called SignalFx. So Splunk acquired our company last October, and we had built a set of observability and monitoring solutions for operators of modern cloud applications. Since then, we’ve specialized more on the modern stack. So now as a part of Splunk, we are the cornerstone of Splunk’s observability portfolio and focus on the DevOps, AppDev buying centers. So my responsibility at Splunk is really to make sure that, cross-functionally, all of our efforts are aligned and focused on the right things, particularly as it relates to observability and application monitoring.

Daniel Newman: So I have to ask you, because we have a wide swath of listeners, a lot of industry, a lot of technology leaders, I wrote about SignalFx being acquired. Pretty cool. Had to have been a pretty big day, or, obviously, it didn’t happen in a day. The announcement happens in a day, but that’s pretty big. That’s a pretty big exit. Was that exciting, having Splunk come to the table? What was the feeling like to have that moment?

Karthik Rau: It was really exciting. It was an exciting process, and it was an exciting day when we announced the deal. It was almost a year ago. End of August was when we announced the deal. So, as you said, obviously, there was a ton of work that went into building the business and a ton of that we’d spent with Splunk in the months prior to the acquisition being announced, just in negotiating the deal and working through that entire process. But I remember the day that we announced the acquisition, there was just this excitement within the company.

We were in all-hands, and we got everyone together at 12:45, because the press release from Splunk was going to hit the wires at 1:00. So we wanted to tell everyone just as the news is being announced, and Splunk had some people there just to kind of record just on the background without anyone knowing.

But it was super exciting. Everyone was up and cheering and excited, and there really couldn’t have been a better fit for us as a company, if we were going to be acquired, than Splunk. Just culturally, it was just such a great fit. Product portfolios just fit in so seamlessly, and Splunk is very much a young company still. So we felt right at home from the very beginning. So truly an exciting day and exciting process for us.

Daniel Newman: Yeah, that’s awesome. Congratulations. One year at Splunk now. I just have to imagine that room. Any time a small, fast-moving company calls an all hands, everybody knows something’s up, because small companies don’t have meetings for the sake of meetings, and you weren’t small, but you know what I mean? Startup-type companies, there’s no meetings for the sake of meetings. Every meeting means something when you’re in that phase. So that’s really exciting.

Speaking of a year ago, wow. We’ve seen a massive change over the last year. Our world is completely different. In your role, you’re watching that really closely, specifically with regards to IT. Just kind of curious for your perspectives on how has COVID-19 impacted what you’re doing, but also kind of remote work, how it’s changing the way IT teams are working together. I’m kind of interested, because I know you’re very involved in everything from development and deployment of microservices, DevOps frameworks, all the way to traditional IT management. What are you seeing? Kind of what’s the shift look like in this past few months?

Karthik Rau: The world has definitely changed permanently, I would say, in the last few months, and, in my opinion, basically what’s happened is it’s accelerated by about five to ten years what would have happened. It just happened a lot faster. At a macro level, I really believe that the triggering event to all of these digital initiatives and the changing kind of landscape of IT was spawned in 2007 with the smartphone, right? iPhone comes out. Pre-iPhone, there are a few hundred million people connected to the Internet, and they’re connected when they’re sitting at their desktops. Today, there are five billion people connected to the Internet, and they’re connected all the time. So minutes of software and Internet consumption is up through the roof.

What COVID is doing is it’s accelerating that even further. So while the number of people connected to the Internet is kind of beginning to get to that plateau where it’ll continue to grow and creep up to 100%, minutes of consumption is just that even more so in the last six months, right? So every company now is forced to be a digital company, whereas before, maybe they could have taken a little bit more time. Now there really are no other options. If you don’t have a digital strategy and this continues as it has been or relapses to the shelter in place, you’re screwed.

So what we’ve seen is a rapid, rapid, rapid acceleration of digital initiatives. We’ve seen a rapid increase in the adoption of SaaS services, because now, of course, almost all workforces are a lot more remote than they used to be. These were all trends that we’re headed in that direction, anyway. It was just accelerated by five plus years just because of this pandemic.

Daniel Newman: I think that’s so interesting. I’ve said we sort of saw a decade of digital transformation in the matter of a few months. It kind of goes to show there was two things that I really saw, was one is sort of the great divide of the enterprise haves and have-nots, in the sense of companies that knew how to change were able to literally put their foot on the accelerator and just explode. Kind of made me wonder why they didn’t do it before. But at the same time, I think it’s manageability. It’s innovation. It’s innovation at scale. We all know the law of diffusion of innovation. You don’t want to roll out the great next product while you still have a huge up cycle with the current product. But at the same time, you’ve always got to be careful to not let that path cross where you get disrupted because you wait too long to come out with things.

But then you also saw on the other side companies that really had sort of sat on their laurels, sat on past success, past business models that were highly based on “analog models,” just how destructive it was for some of these businesses. I think every company on the planet that’s survived this, whether it’s by hook or by crook, is going to think differently, going forward. One of the big think differentlies has been the shift to cloud, too. I have to imagine even for Splunk, your business model changed, and a big part of that happened when you came onboard, around that time when the company really shifted its model to more consumption-based, more cloud-based, because while it is all about observability and seeing all the data, the model that Splunk went to market with wasn’t traditionally as “cloud” as it is now. How much do you think this has accelerated that model, and how much do you think that’ll last in the longer term?

Karthik Rau: Well, I think it’s accelerated a trend that was already inevitable, right? Just like everything else, the move to SaaS and the move to cloud was already kind of headed in that direction, not just because of the simplicity of the deployment models and the time to value benefits of SaaS, but nowadays there’s so much more you can do, that a provider can do in terms of data analytics if they actually have the data running in their cloud service, right? So data analytics is getting more and more powerful, and customers now appreciate that if they’re leveraging the cloud service, there’s so much more that their vendors can do for them.

So that was an inevitable trend. I think to your earlier point, it’s really great to see organizations that have got their digital strategies in place, how well they’re doing and how they’ve accelerated it when they needed to. I think if this pandemic had hit three or four years ago, a lot of businesses would have been in a much, much, much tougher position, because, really, in the last three or four years, there’s been so much momentum towards cloud, so much momentum building digital initiatives, having an online presence, having direct to consumer-type applications. So we’re in a far better position today as a world and I think just as an economy than we would have been a few years ago.

Now, there are certainly some people who are late to the game, and I think they’re being disproportionately affected here because they haven’t got their digital strategies in place. But fortunately, I think most people have some semblance of something in place today.

Daniel Newman: Yeah, absolutely. That makes me think you sort of mentioned cloud applications and SaaS, but what we’ve actually really seen, at least as an analyst that looks across the tech stack, is hybrid. We’ve seen that, by and large, the big enterprise and even the midsize enterprises are really sort of putting their chips down on a hybrid infrastructure. There’s a lot of reasons for this, everything from data protection and sovereignty, security to regulation compliance. Also, you’ve got control. You’ve got the ability to be flexible. You want to be able to run workloads in multiple locations quickly, which is pushing hybrid and multi-cloud, but this has also kind of created a renaissance for IT.

Over the last few years, we’ve heard a lot about more tech being bought in the line of business, these tools, but those are appliances, in a lot of ways, to a business. As companies are trying to build a stack of software that’s unified, you’re hearing more about what you guys are doing. You’re hearing more about platforms, low code and no code, stuff that lives on a hybrid environment that can touch the ERP and CRM and HCM and unified and … Have you noticed that? Because, to me, it kind of feels like, you guys kind of being the overarching monitoring and management of all IT, I feel like there’s this renaissance where IT has gained a new level of importance that was sort of starting to be pulled out from underneath during this period when we thought public cloud and SaaS was going to win. Now it’s really about hybrid multi-cloud and hybrid software deployments.

Karthik Rau: We’ve always believed and I’ve always believed that IT’s role just evolves with this new model. Anyone who had said that IT disappears or you go to kind of a just pure developers running everything, we never saw that role really emerging. We’ve been focusing on the new stack. Now, the role changes. It certainly changes from … What we’ve found is it’s really about providing a set of paved roads for these development organizations and end users, in the case of more classic business apps, so that they can effectively … If it’s a developer, they can move very quickly without breaking things, right? So you have all of the infrastructure set up. You’ve got all of the tooling set up. You’ve got all the monitoring set up. You’ve got all of that set up so that the developers can effectively focus on what they do best, which is developing software, releasing software, and moving very, very quickly.

So there’s still a significant need for that centralized team to provide that set of best practices, that set of kind of core platforms for all of these different end users to be able to leverage. So we are seeing a renaissance of IT in that sense, that the role is changing a little bit, but we’re seeing them come back to the central focus again.

Daniel Newman: Please, by no means was I belittling. I was just saying when you started talking about broad transformation, there was just studies out. There were studies that digital transformation was going to be led by the CMO. I mean, there was an absurdity to it all, in my opinion. Now, again, I think it’s cross-functional, and I think that’s the one thing a lot of people miss. Culture and people are drivers as much as technology. So you have to have experiences baked in. But when it comes to the tech in the stack and the data and the AI and the ML and the things that are going to actually make these experiences possible, IT was largely going to be at the core of making this possible.

But there was just so much energy being put into these off the shelf, SaaS tools that you could import your data from your Salesforce administration and pinpoint offerings in front of people by using retargeting and stuff like that. It’s like it was a microcosm. It was something that could be done in isolation. But when you’re talking about a bank or a healthcare system, the idea of being able to do that in isolation from IT was crazy, but I’m telling you, I heard about it all the time.

Karthik Rau: When you look at the grand scheme of things, right, how much organizations are spending on what would have been packaged applications, which now much of that is delivered by a SaaS, versus what they’re going to be spending on their digital transformation initiatives that are really revenue drivers for their business, what they’re going to spend on their digital initiatives is going to dwarf kind of what they had spent on their classic business applications, because it’s really what … Every business is becoming a software business.

So if you think about it that way, IT’s responsibility really becomes how do you enable your organization to build the software that they need to drive their digital businesses forward as quickly as possible? That is a significant role. Historically, the way software development used to work is you’d have these big release projects, and you’d have once a year updates of your packaged software. But when you’re talking about all of these revenue-generating digital apps today, companies need to act more like Facebook and Google, right, where they’re releasing software daily. They’re just out there and competing very effectively against these small and nimble competitors.

In that world, you really need to think very proactively about how you build your engineering organization. How you equip them with all of the tooling? How you equip them with the data so that they can move quickly without breaking things is kind of how we like to talk about it, and that’s a huge, huge, huge need for every organization if they want to be successful with their digital initiatives. That’s really what we see as the big focus of kind of what IT is going to be over the next 20 or 30 years, is enabling their organizations to do that effectively.

Daniel Newman: Absolutely. I think you guys are a great example with everything you’re doing. So I’ve got a couple more questions, a little bit more time with you, Karthik. So thanks for sticking with me. I appreciate you and Splunk taking the time. Let’s talk about industries. We talked kind of holistically about what’s changed, what’s accelerating, what’s moving. What are some of the industries that you’ve really seen, if you had to pick, that you’d say, “Hey, these industries, more than any others, in the wake of COVID-19 have seen acceleration,” especially as it regards to IT?

Karthik Rau: Well, several, right? I mean, I think education is one that has very clearly been affected by this. There’s a significant dramatic shift to remote learning, right? Remote learning, that’s affected, I think, every educational institution. Healthcare we’ve seen moved towards telemedicine or kind of a need to get their telemedicine initiatives and kind of regulatory standing done overnight. Public sector, again, just you can’t go in-person to the DMV, as you used to. So I just got my driver’s license renewal. I’ll have to do everything online as much as possible. So all of the public sector sorts of customer-facing or citizen-facing applications that needed to move online.

So we’re seeing it in a lot of places that, again, maybe would have seen it in a few years or been forced to do it within a few years having to do it and on the order of months. Now, that’s kind of on top of what we’ve generally been seeing across industries, where, as a byproduct of all the digital transformation, every business has had to transform itself, right? Consumer goods companies have had to go direct to consumer, directly from them. That’s been a major initiative across a number of industries. A lot of organizations have had to move to mobile first, right? Banks have had to invest significantly in all of their mobile applications. Media has been investing quite substantially in streaming.

So pretty much you can pick any industry, and there’s been a massive transformation. Reinsurance, I was just talking to a CIO of a big reinsurance company, and their customer base has fundamentally changed, because it’s not just the big insurance companies anymore, but a bunch of small online emerging insurance providers that love to use APIs in the backend versus the more traditional means. So we’re seeing it everywhere.

Daniel Newman: Yeah. FinTech is going to change the course of almost everything right now. It’s interesting you say that, because I do think there are certain examples that come to mind for everybody. I was thinking to myself ride sharing is an interesting thing. It was exploding, and the technology is fantastic, but it’s been completely disrupted by the fact that people don’t want to be in random people’s cars. You think, “How fast is this going to accelerate self-driving?”, because people still want the ability to get a ride. They just aren’t going to necessarily want to be in someone else’s car. So, I mean, you could literally pick anything. You could pick anything right now and say, “It’s going to go fast.”

I do think we still have issues. You mentioned public sector, healthcare. These are all industries that have a humongous need for transformation. I don’t know about you. Do you have kids, Karthik?

Karthik Rau: No, we don’t have kids.

Daniel Newman: No? I do, and the only reason I point this out is I have kids in school that were in high school and in junior high during this, and the education system was entirely unprepared. I mean, we live in a nice area with good schools. The online schooling, when they went home on Friday and came back on Monday, it was completely unprepared. But it wasn’t a lack of purely just technology preparation.

Culturally speaking, these industries have to completely rethink the way they design curriculum and develop. Telemedicine, this has worked forever, the idea of being able to get on the phone and talk to a nurse or a doctor, get a prescription. Again, there’s so many of these things that the tech isn’t the problem. The culture is the problem.

So you sometimes see in some industries the tech can drive transformation because the industry is very willing. Then there’s industries that have a really big resistance, that you can have all the tech in the world. Public sector is another great example. Gosh, I remember I was talking to an agency as a business owner, trying to get a bill paid, and they were telling me they could not send me a piece of information because they’d all been sent home from work and they had no way to securely send me a document, none whatsoever. I’m talking about the people that are responsible for federal taxes. Let’s just say it was that group, and I’m saying this to think, “We can’t even get you a document,” because the only way they could send me something, Karthik, was fax.

I mean, it’s crazy that we are in 2020 and people cannot … The systems are so broken, because, of course, we’re on a phone. You had a computer. You had email. You had a million different ways to potentially send. But the one thing a lot of people don’t have in their home office anymore is a fax machine.

Karthik Rau: Right.

Daniel Newman: I mean, we got rid of that. So let’s take it home on this question here. Observability is a big part of what you do and what you did at SignalFx, and now what you’ve brought to Splunk was another level and another advancing the capability. How are you seeing that being applied, utilized, changes in behavior, anything as it not only is a result of what’s going on with the pandemic, but in terms of just this whole driving digital transformation? Where’s this going? What are you seeing in terms of companies making greater investments in being able to really observe the data across all of their IT?

Karthik Rau: Well, we’re seeing the biggest set of changes to how applications are built, deployed, and run since people move from the mainframe to distributed systems in this sort of period of digital transformation, and the key drivers for that are the underlying infrastructure has changed from on-premise data centers to highly elastic public clouds and private clouds. The application architectures have changed from these big packaged software applications to these bite-size little microservices that are front ended by APIs. For example, when you check out on Amazon, you’re probably hitting hundreds of microservices behind the scenes, and the reason for that is you can have small teams that are working remotely and independently. They can update their individual service, and all of the different services can continue interacting as they were before.

So that’s a huge change in application architectures and development process, and then the operating models have fundamentally changed because people are now delivering software as a service. So you’re not just shipping packaged software and having someone else run it. You’re running your own software. So that drives all of these DevOps models, where you build it, you run it, and developers have our vested interest in supporting their own software.

So these are three fundamental architecture and people process changes that we haven’t seen again since people moved from mainframe to distributed systems. So what that means is, at least in the world of today’s applications, there’s orders of magnitude more data that could be generated, because there’s so much change. There’s so many people involved. There’s so many systems that in order to make sure that you don’t make a mistake and you don’t bring down your application for all your users, you need to capture a lot of data on every single change that’s happened in your system. That’s orders of magnitude more data than you ever generated before.

So in addition to generating all of this data, you need to have systems on the backend that can process all that data very quickly, correlate trends across all of these different teams and infrastructure components, and surface the patterns that are interesting to you quickly enough so that if you do make a mistake, you’re able to roll it back very quickly and undo it.

The secret to success in this new world is you can’t avoid mistakes. You just have to make sure that when you make a mistake, it’s small and isolated and you can roll it back as quickly as possible. So when you have that safety net, you can move as quickly as you want to and feel very comfortable being aggressive with your software deployments.

That requires an observability solution. It requires a solution that can instrument and collect data everywhere. So telemetry data, think of all the telemetry that gets collected on airplanes when they’re flying. You’ve just got to collect as much telemetry as possible and as much depth as possible across your entire application and infrastructure landscape. Then you need a system that can process it very quickly, and that’s what we do at Splunk. We’re the data to everything platform. We have historically captured any kind of unstructured data, and now with the combination of the SignalFx and Omnition, which is another company that Splunk acquired, with our technologies, you can capture a bunch of telemetry data for metrics and traces and have the best backend engine for processing all this data very quickly, identifying the trends, leveraging the right AIML to surface things that are interesting to you.

So by using our solutions, you can basically have the safety net that you need to move quickly without breaking things. That’s in a nutshell what observability is and what we do differently at Splunk.

Daniel Newman: That’s a big and important investment. I think a lot of times, like I said, they’re looking in microcosms. They’re looking at specific activities with data, and the thing about it is there’s an entire data ecosystem. I wrote an article that just published, depending on when you’re listening to this podcast, but it’s August 2020. It still feels like March for everybody. But I wrote on Forbes, I do a trends piece every year, basically top ten digital transformation trends. One of the trends is not really all that visionary from last year, but it talked about the continued pervasive investment that’s going to happen in analytics because of data, the sprawl of data.

One of the things I actually said about Splunk that was really interesting was last year, Splunk came out with that data to everything moniker or that slogan or campaign. I said I think a lot of people sort of looked at that slogan as a little ambitious, a little broad. I said we’ve come to realize now after what’s happened with COVID and the need that businesses could really be able to quickly, successfully wrap their arms around, to manage, ingest, nurture, enrich, and use the data in real time, I said there was quite a bit of foresight there. I said there was a quite a bit of foresight there that maybe the company didn’t get enough credit for.

So, again, it’s very interesting to see how this wrapper around all your data, to be able to really run your business and make sure your IT is up and running is going to be so important.

Karthik Rau: I agree 100%.

Daniel Newman: So Karthik, I want to say thank you. Appreciate you putting up with me. I mean, spending this time with me. I mean, enjoying this podcast with you. It was actually a lot of fun. You’ve got a tremendous handle on it, and I’m sure that’s a big part of why Splunk made the commitment to acquire SignalFx. They always say the founder is such a big driver in any of the decisions. So congratulations on that success. Congratulations on what you’re building now at Splunk. Really appreciate you joining me on the Futurum Tech podcast today.

Karthik Rau: Thank you, Daniel. It’s been great to be on.

Daniel Newman: So for everyone out there, go ahead and hit that Subscribe button. We really appreciate you being a part of the Futurum community and the Futurum Tech Podcast. We appreciate Splunk as your partner for being part of this series. Check out the other ones. We had Doug Merritt, the CEO, as well as Tim Tully, the CTO, and now we’ve had Karthik here. Great series of podcasts. A lot more to listen to on our channel. So, again, hit that Subscribe button. Appreciate your commitment to being part of our community. But for now, for this episode, we’ve got to say goodbye. We’ll see you later very soon on the next episode of the Futurum Tech Podcast.

Image Credit: Yahoo Finance

About the Author

Daniel Newman is the Principal Analyst of Futurum Research and the CEO of Broadsuite Media Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise. Read Full Bio