RESEARCH & RESOURCES

Q&A: The Next Generation of the Internet of Things

The second generation of the Internet of things combines processing concurrency along with scale-out architectures to enable real-time decision-making, explains Monte Zweben, co-founder and CEO of Splice Machine.

Monte Zweben is the co-founder and CEO of Splice Machine, which offers the Splice SQL Engine, a massively scalable SQL engine for big data applications. "The Splice SQL Engine delivers the scalability of what NoSQL databases strive to deliver while providing the secondary indices, joins, and transactions that are critical for Big Data applications."

In this interview, Zweben discusses the coming second generation of the Internet of things (IoT), in which scalable architecture and concurrency will come together to enable us to move, as Zweben says, "from [only] consuming data from the Internet of things to actually controlling the Internet of things."

BI This Week: How is the Internet of things changing data management in basic ways?

Monte Zweben: The first thing is the volume of data. You have a plethora of devices that are suddenly online, putting out data for enterprises to consume, digest, filter, analyze, and take action on. It's an incredible burden on existing data infrastructure because of the sheer volume. That's being talked about quite a bit.

What has also been discussed is the velocity of that data. No longer can we take data from a point-of-sale machine in a retail setting once a night. What we really want is have every transaction communicated online in seconds or less. For example, the ability to detect fraud at these devices needs to be done in milliseconds. Similar [uses] across the supply chain mean that both the volume of data managed and the velocity are absolutely huge.

That's not news, though. What is new is that we are crossing into a new era regarding how to manage this data. The first generation of the Internet of things required data infrastructure to do the first part of that equation -- consuming data, digesting it, storing it, and making it available.

The second generation, which is what I'm focused on and excited about, is about building applications on top of the Internet of things. How do we take care of data and at the same time control and interact with it in real time? That requires a new approach. So that's my passion – moving from just consuming data from the Internet of things, to controlling the Internet of things.

How far along are we regarding that second generation that you describe -- building apps on top of the IoT?

I believe we're very early. I think the first generation is well underway -- people are using a variety of scale-out architectures to consume that data we just talked about. Those scale-out architectures were the enabler -- they were the architectural breakthrough that finally enabled many businesses to take in a high volume of data at high velocity.

By scale-out, I mean that instead of trying to store all that data on a single big computer, with a lot of memory storage and network bandwidth and CPUs, we were able to take that data and spread it across [many] inexpensive commodity machines. That new scale-out architecture is what generated the capability for Generation One of the IoT.

The second generation requires a new architecture, one that the old architecture always had -- data platforms that allow multiple people and devices to be interacting concurrently. Now we need to apply that architecture to the second generation. If you have lots of people and things changing your data all at the same time, you need a different kind of architecture, and that's where the second generation is going to come in.

That's where we spend our time at Splice Machine -- dealing with the concurrency that's required for powering applications. To answer your question, I think we're still in the early days. The companies that are supporting this transactional concurrent architecture -- companies such as ours -- are in early inceptions, with a few customers proving it out.

We're going to see an explosion in the next few years of use cases as the technologies are proven and rolled out -- the combination of concurrency and scale-out architectures.

Speaking of use cases, can you give some examples of what we might see in a few years as this really takes off?

Supply chains are one scenario. Imagine trucks and supplies being tracked throughout the supply chain. Wherever they are moving, we'll know about it in a very granular and timely way. Even more important, applications will be rapidly re-planning where that product, those trucks, and those containers need to be.

My first company was a supply chain optimization company back in about 1993. We became the heart of PeopleSoft's manufacturing solution. We built an artificial intelligence planning system, but the data that was required simply couldn't fit on single boxes. It wouldn't work at enterprise scale, so we had to scale back and do it departmentally.

Now, however, we can work at enterprise scale, and the holy grail of dynamically getting the right product at the right time to the right person and being able to both forecast and execute the delivery is coming to fruition. The key is having everything tagged and knowing where it is in the supply chain.

[That's just one example] of the kinds of things you're going to see in real time. The data doesn't just get stuffed into a big database with reports coming out later. You're going to see real-time decision-making, and applications that are changing behaviors in the enterprise.

That's a good example at a somewhat granular level. What's the big picture with the Internet of things?

Imagine total automation. As time goes on, imagine if everything you own is literally on the Internet -- your thermostat, lights in your house, your toaster, your car, your garage, the locks on your door -- when your very own "Internet of things" knows where you are and where your family members are and can learn from that. I think it's going to be a different world, an automated world. ... That's an example of the big picture.

TDWI Membership

Get immediate access to training discounts, video library, BI Teams, Skills, Budget Report, and more

Individual, Student, & Team memberships available.