Podcast Players Logo
Ravi Mayuram

Innovating International Growth

Special Guest

Ravi Mayuram

Chief Technology Officer of Couchbase

Episode 7

June 20, 2019

share icon

Share

Innovating International Growth

If you think your platform is ready for global expansion, then you’d better make sure your technology can keep up. Today, Couchbase’s Chief Technology Officer, Ravi Mayuram, breaks down exactly what you need to succeed in new geographies.

Transcript

Announcer:

Do you hear that? That’s the sound of platform innovation at work.

And this? Is PLATFORM PLAYERS.

If you think your platform is ready for global expansion, then you’d better make sure your technology can keep up. Today, Couchbase’s Chief Technology Officer, Ravi Mayuram, breaks down exactly what you need to succeed in new geographies.

But first:

This is your Platform Player Flashback!

Narrator:

Welcome to the Couchbase office.

Yeah, thanks for meeting with us. We have a big problem on our hands.

Another nervous CEO, flanked by his entourage, walked through the glass doors to the conference room and tentatively took a seat.

Can I get you some coffee?

It was just another day at Couchbase.

Ah – here’s our CTO, Ravi Mayuram.

…I’m sorry, how did you say that?

Ravi. Ma. You. As in “Y-O-U”. Rum. As in “Rum and Coca Cola.”

Ravi Mayuram. Thanks for meeting with me.

Ravi took a seat and gazed across the table at the CEO, who was fidgeting with his cellphone.

What’s the trouble?

We’re scaling massively. We’re just launched in Europe and Asia. We have over a million users.

Great.

But all of a sudden our platform is slow. Our users are getting “the beach ball of death” every time. We don’t know why.

It’s because your data doesn’t live in the geography you’re serving.

We’re losing them.

That’s a problem, yes. You need an NOSQL database. We also need to make sure that you’re meeting the geographic regulations.

I didn’t know there were geographic regulations for platforms…

There are.

The CEO unraveled. But Ravi stayed cool as a cucumber. This was a conversation he had several times a week.

Can you help us?

Yes. Which was the answer he gave several times a week. Would you like a coffee?

Do you have anything stronger?

Announcer:

FLASH FORWARD! Couchbase has raised $155 million in funding and became the go-to enterprise database platform, servicing clients like AT&T, Cisco, Comcast, Disney, DreamWorks Animation, and eBay, as well as hundreds of other household names.

Let’s get down to the NITTY GRITTY. Here’s your host, Kurt Bilafer!

Kurt:

Platforms have the remarkable ability to transcend borders. And if you look at industry trends, you’ll find that more and more of them are “going global.” This often comes with a set of unforeseen technology challenges, as you’ll learn from today’s expert guest. Ladies and gentlemen, join me in welcoming Couchbase’s CTO, Ravi Mayuram. Thanks for being here, Ravi.

Ravi Mayuram:

Thanks, Kurt.

Kurt:

Of course, Ravi. It's exciting to be here with you because I know a big part of what Couchbase is focused on is helping a variety of companies, not just marketplaces, set up your infrastructure appropriately. Especially as they start expanding internationally.

That's really the angle I wanted to take today was to understand what are the sort of decisions, tradeoffs, thought processes that marketplaces need to go through as they start thinking about processing and growing their business outside the U.S.?

Ravi Mayuram:

Yes, it's a very interesting topic, and it is something that comes with very often the boundaries which we used to have because of physical limitations are gone now because if you launch a service and it is popular, the next day it is global in scale instantaneously.

But with that comes concentrations that you should be aware of which part of it is government regulations. Part of it is logistical, and some of it is straight up user experience because if you had built with the part of everything being centric to the place we are from – if you were from the U.S., you wouldn't realize the latency in regulations if you were to do business in a country like India or somewhere in Europe.

One of the first considerations is about the data privacy, GDPR oriented concentrations. Now, even if you understand that, how do you implement that? Some of the older technologies are not built for modern-day fluidity of the business.

That's where Couchbase comes into play on many occasions to give people those infrastructural capabilities in the data platform that we have where people can geo-distribute the data in a very easy and on-the-glance matter as opposed to hiring consultants, and programming, and taking month-long projects or year-long to implement this.

Now, with this modern platform, you can do that very easily by clicking a few buttons and setting your priorities and preferences, and the system takes care of the rest for you. In this world, what we specialize in, one of the areas is also the geo-distribution of data and geofencing of data as well.

So, this gives you an opportunity to lay out your policies in terms of what data can move across geographic boundaries and what cannot and what needs to be staying within a country. You can set those policies and the rest of the stuff the system takes care of.

So, even if you start with thinking of just one geography and the business suddenly requires you to expand to other geos, it can now be done by fluidly extending this infrastructure across the globe. You solve multiple issues by thinking in a geo-distributive form.

You solve for performance because people who are, let's say, suddenly if you want to enter Australia, people should not have to come all the way to the U.S. to find that data. It should be located somewhere close to Australia so you can get to that. These are user experience sort of concentrations.

Then comes the consideration about redundancy of the data or there so the application is always on no matter if a certain availability zone goes down, still the data should be available. That can also be achieved with these technologies.

Kurt:

So, Ravi, that's great, and there's a lot in there. Let me try to unpack it. One of the things you talked about was just the five nines. So, 99.999% uptime, which is usually like an SLA requirement.

Ravi Mayuram:

Correct.

Kurt:

For online services, I think one of the issue things is depending on what size company you're talking to and where they are in their journey. Oftentimes, they're just like, "I'm just trying to keep the lights on." There isn't often a thought of like, "How do I make sure I have this consistency?"

Are you seeing rules and regulations across the globe change? There certainly is a change from customer expectations. Oftentimes in our podcast, we talk about these divinely discontent customers where once you satisfy them – the rule used to be you had eight seconds on a mobile device, and it's no longer eight seconds.

Ravi Mayuram:

Absolutely.

Kurt:

When do you think these companies actually start thinking about this? I'm sure most startups are just saying like, "I've got to throw something together and use a basic cloud infrastructure whether it's Microsoft, Amazon, Google." When do they start thinking about "How do I actually make it more professional, scalable in maintaining those requirements?"

Ravi Mayuram:

You know, that's a very interesting question. On a different zone, the DNA of the company and sometimes, industry. But generally, it comes down to this: when you build something, you are in this mindset of if it is that successful, let's solve it then. If it is that big a problem, then let's get to solving this.

But it is becoming a little bit more – how do you say architect-driven as opposed to just development-driven these days where these concentrations are thought about. When picking a platform, even if it is not implemented, they've won that optionality in the platform that they will not be able to think without a paddle.

So, they make the technology choice of what to pick, even if they haven't implemented the solution to the extent that they want to. They want the platform to be one that will allow that so that they're not logged out of it. That they don't have to go back and rethink their whole strategy eventually.

So, it's a little bit later in the maturity of the company. As you pointed out in the beginning, perhaps it is not that critical because they don't know there's going to be that success to begin with. "If we're successful, yes, we want to solve it. When the time comes to solve it, the technology decisions I'm making now will allow me or will preclude me from doing that" is the judgment call they have to make.

That's when we get engaged with them, and we have conversations and help them understand the problem because we have many other companies who are further along in this maturity who have actually perhaps made some mistakes and then come to us and work with us.

Or, perhaps, they chose us, and we learned from them in terms of how these problems need to be solved. So, it's a combination of maturity, successes, and the industry that drives at various stages where they need to solve this problem.

Kurt:

You pointed out that oftentimes people just defer these decisions to like, "Once we are successful and prove this works," we have often called those champagne moments. But the reality is with the regulatory environment that it is today, that's no longer good enough.

I mean, the reality is that if you aren't GDPR requirement and you have users in the UK, you could still be vulnerable. So, you actually need to start thinking about these things earlier in the process. At least, have a mitigation plan in place to understand what you're going to do if you start seeing users in these geographies.

Ravi Mayuram:

Absolutely. It goes further. There are other compliance standards also. Like in the U.S., there are standards for if you are holding someone's personal information, there are PCI standards. The health care industry has it's on HIPPA standards.

So, you need to understand both of these right at the beginning as you point out whether you build it out or not, that could be the choice you want to make eventually, but you need to make the technology choices early on which will allow you to implement those regulations in your applications when the time comes. Not pick a stack and then regret that "Well, now I can't do that," and then you have to basically go back to the drawing board which is a pretty expensive process.

Kurt:

You're right. We've been talking primarily about the regulatory angle, but the reality is, from a customer experience standpoint, if someone is trying to use your solution overseas and has a horrible experience, one or two things are going to happen. Either they're going to stop using it – first off, they're going tell everybody how bad the experience was, which is never great.

What seems to get lost on the internet is that location is relevant when using these services. I think most Americans and certainly U.S. companies have no understanding that 4G doesn't exist in every major metropolitan area. We work on multiple screens where lots of people just have one screen, which is primarily a mobile device. So, I think there are a lot of different things to consider.

Ravi Mayuram:

You're absolutely right. It's what we call geo-location of data is what's important. Actually, you bring up an interesting point there, which is about the edge. If you see all the mobile applications, there are many of them where you would actually see that you're hounded by the circle of debt, as they call it. You clicked on something, and it's going to be a few seconds before actually your application refreshes or you can actually do your next act on it.

Mostly it will be blamed on the network. In fact, it's not the network that's to be blamed, the fact is that the data that you wanted to operate on is not on the device. You're doing the application logic of the mobile device on the mobile device. It's running locally. The data is actually being fetched somewhere far in a data center someplace which is behind a couple of firewalls, and God knows how many hoops it is going through.

But what you really need is that data that you're actually going to be working on should be at the edge on the device so that the application is always on. Irresectable of the network band width availability or dark zones that you enter and stuff like that.

So, this is where the user experience is going because more and more of this stuff is actually happening. They have established that 60% of all e-commerce actually these days is the mobile devices and stuff like that. So, the point of engagement is a mobile device, and the data that you're operating on needs to be on that device. How do you distribute the data is a huge strategy you need to think about in the beginning?

Kurt:

Where do you get started? I'm just imagining you're two guys because that's typically what it is whether it’s the right mix or not. But it's two guys.

Ravi Mayuram:

Two guys in a garage?

Kurt:

They come with a whiz-bang idea, and they go off, and they start building. If you read all the books, it's all about this minimum viable product if you don't lose early, or lose fast. So, I wonder, like where do they start thinking about this, because the examples we've been talking about are you're going to another geography. But if you've built a great application, whether it's mission-critical or it's just something that people are using frequenting, when they travel abroad the same situation is going to arrive.

So, this isn't just like, "Now I'm going to be going after UK customers." This is, "If I'm really building an application that people are using globally, how do I make sure that they have this desired experience?" Who's job is that? Is it the CIO, the CTO, the head of architect? It sounds like, depending on the stage, it could be a variety of different people.

But who's really thinking about this, and how early in the process do they start drawing up plans and say, "How do we do this?" Because as you said, it's becoming table stakes and hopefully, companies are starting to build themselves more for success now. So, I'd love your perspective on that.

Ravi Mayuram:

Yeah. I think it's a combination of people or market situations that drives this. From a technology standpoint, definitely, it's the architect who has to think through this. The other person concerned is a little bit less sort of global if you will. They have to think in terms of the feature functionality, performance, and the fact that it is sort of nothing is going to break or edit out in front of the customers. That's what they should be concerned about to make sure that every individual user has a great experience.

The architects have to think systemically as in "Okay. Now that we've put this one out, how do I now make sure that it’s built into this, so that no matter what, the application is always on which is the 5/9 availability you talked about earlier.

What is our strategy to achieve that? Well, if my lovely little desktop underneath me is where I'm running this thing, that isn't going to cut it. Maybe I need to have a presence in the cloud, and then if it's in the cloud, then what if that cloud provider fails, their one data center fails, do I have multi-data-center strategy available at these zones that exist in the cloud.

Maybe I need to have my service running in more than one availability zone. When you do that, you get into the next question of what is the elastic capacity that I need? Tomorrow, instead of just ten users showing up, ten million users show up. How do I cater to those 10 million users with 100,000 users in the system? How is my performance going to be?

So, someone needs to think through these set of problems and have the architect there be robust enough. It's a constant evolution. But day one, you don't have to think about this like a new arrival product has to prove the business value that is valued in this. But once you grow the value, when you take it out, and you start deploying this, and the customer starts to walk, and it's basically, you need to think about having 50 stores across the U.S. if you're just starting one store on 1st and Main? When that thinking about the second store or the third store, when do you need to have a strategy around where to open stores, it comes as a natural evolution once the application of the service becomes that much more successful.

So, there's time as part of innovatively and as part of the growing the business that you actually think about these problems in various stages. But there are a few foundational ones that you have to think in terms of the elastic capacity, the performance, and sort of redundancy, and availability. These you have to think right up at the beginning.

This is what's happening these days with the modern applications because you can take a whole bunch of services, and the cloud's elastic capacity, and you can start from there. So, the modern architecture started at that standpoint of two things are given.

Now, let's make sure that our service can take advantage of what the elastic capacity, the geo-availability, and distribution that is available in the cloud, how can we leverage those services in our application so we provide the user experience as for last availability to be compliant wherever we're going to be running this service?

Kurt:

What's the best piece of advice that you've ever gotten?

Ravi Mayuram:

Best piece of advice I've ever gotten. Hum. Always stay a student. Don't try to be a master.

Kurt:

Who provided that advice to you?

Ravi Mayuram:

The gentleman who I go and learn drums from also mentioned that as part of the teaching lessons. This was his general advice that he got from his father.

Kurt:

Okay. Cool. What's your favorite smell or sound?

Ravi Mayuram:

Smell or sound. Wow. The smell of the ocean is my favorite smell. Somehow it seems to indicate to me the vast expanses that are in front of us and the possibilities there are. Sound-wise, I would say maybe it comes together at the sound of the waves.

Kurt:

What would you be doing if you weren't doing this at Couchbase?

Ravi Mayuram:

That's a very good question. I don't know. Maybe I'll take the drums more seriously, or maybe I'll be playing golf more often.

Kurt:

Good. Good. The last question is, what do you still enjoy doing the old-fashioned way?

Ravi Mayuram:

I still like to buy the newspaper and read it in the morning. Sports sections and puzzles are perhaps where I spend most of the time. It's the physical act of having something even though I can find all that in my iPad and have the mobile version of these applications. Somehow, the physical act of flipping a newspaper to go read or find, there is still excitement in that.

Announcer:

Time's up!

Kurt:

Ravi, I really appreciate you joining us here at Platform Players. I think the interesting thing is what seems to dominate the headlines in most things we read are the successes of these individual companies. But what's often missed is the ones that stand behind them, that enable them to do that. I know you guys play a big role in that. So, again, thank you so much for joining and sharing that with us today.

share icon

Share

Latest Episodes

EP. 10

Matthew Hoffman

Division President of Kigo

Latest Episodes
Podcast Players Logo
© 2019 Yapstone, Inc. NMLS #1488912All rights reserved.