 Welcome back everyone to theCUBE's live coverage here in Las Vegas on the show floor. SAS explores here, I'm John Furrier, Dave Vellante. We're extracting the system of noise with theCUBE and we've got a great guest here talking about edge to cloud, sustainability, energy, IoT. As Dave becomes more important, it's going to have a lot of advantage, not just for efficiency but also for sustainability. Jason Mann, SAS VP of IoT, welcome to theCUBE. Thank you so much. So you got a lot going on. You got products embedded into solutions. You got a lot of customers on the edge. Edge being windmills, you know. Sensor data. Sensor data, weather satellites. Yes. Oil rigging. What's your job? It's pretty complicated. Tell me what you do. My job is making sure that we can transport analytics to the source of information and decisioning. If you look at SAS's 43, 44 year history, we've been dealing with sensor data the whole time. The big change with IoT, and this is a journey that we started internally about five years ago, was to be able to inject ourselves into these ecosystems, get closer to the source of data, make a determination if this is where you need to make a decision and act on it at that point. So that's really been the change. How has been the IoT market been defined? It's just a device with an internet connection and a battery or power. Yeah, it's- I mean, how long can it go? It really has changed a lot on how much you can power up a sensor or device over how much time. That was kind of the start of it. If you go back, again, five or six years ago, our job was really a technology initiative. How can we make our device run in an operating system in an edge device sometimes looking like a router? Sometimes it looks like a camera. So there were many cycles technically to confirm that we could run where we needed to, that we could plug into this existing brown space that we talked about. There's no new installs. So that was a real push. Now that's starting to change a bit. You're starting to see some more of the common initiatives across multiple platforms. We're looking at containers that can work in any environment. You're starting to see standards come about. So it's really starting to change to what are the outcomes we can affect? And that actually goes back to the older school questions. You look at any survey in the industrial sector, there's always three or four things that's most important. There's efficiency. There's equipment efficiency. There's yield. Now we're able to come to them to help resolve those same problems but do it in an innovative way. You know, you mentioned some of the old school techniques, OT and IT and operational technologies. Self-contained Windows 10 machine running something over there that works, hasn't broken in years. So you have these environments that diverse from old tech to new tech. During the presser, you and Brian were talking about some of the use cases. What jumped out at me and I want you to get your reaction to is the concept of how synthetic data and digital twins actually in simulations are going to be a major innovation in manufacturing and then other edge devices and edge environments holistically and also down to the unit, the device level. Yes. I think the synthetic data is just a starting point to discuss that a bit. My background is industrial engineering. So the idea of creating this simulated environment has always been this holy grail. You didn't want to impact the existing manufacturing operations. You didn't want to impact the overall process but it never really worked really well. You know, arrivals couldn't be forecasted the way that it played out in reality. And what we're seeing now is with synthetic data and being able to use generative AI, you can get closer to what these profiles can be and then that feeds directly into the simulation. So then it opens up a whole new opportunity for what if scenario analysis, all that live outside of the actual environments. Is that big, what's changed? Is it because cloud scale and just the timing of what's in the market right now? There's scale and within the synthetic data site just how many parameters that you can feed into these data sets. I mean, the compute power that's there and being able to expand what's impacting the data set has been the big change for us. So what are the economics of IoT look like? How are they evolving and how is it impacting you and your customers? I mean, you've got a different footprint. You've got different energy consumption profiles, cost profiles, performance because you have all this AI and all this data, so you still need very high performance, but it's not traditional computing, right? So how is that evolving and how is that affecting your strategy? Yeah, I think you really see the opportunity play in these high volume areas where oftentimes margin is very small and being able to impact a percentage point or a portion of a percentage point is critical. So what we're finding is this real focus on the importance of making the decision as soon as it can be feasibly made. And I think that's one of the economies of scale that we're getting out of deployments for IoT solutions at the highest level. So I think you'll see that we'll continue to see a push in that. So time to solution, time to resolution. And then the other area is being able to really dig into the processes. Earlier we were talking about some of the energy use. That's a huge opportunity to attack those industries, especially in the process industries that use massive amounts of data and really tune into the granular level of understanding those processes and being able to change their energy consumption profile. And how are the personas evolving? Many came out of like your background, right? Industrial engineering. Is there a developer persona that's emerging that's different than sort of the enterprise developer? Yeah, I think you still have the personas that exist but how they get addressed are different. We talk about internally of the concept of splitting a dev time environment, model development into an operational runtime environment. And we have technologies that can support the decisioning and the models that deploy into those ecosystems. Those are managed and engaged by more of a business user. The model creation happens on the other side of the coin but there's not a daily engagement with the process. They create a model. They run through the model management process. I know you were talking earlier about some of the analytics in the base of that. But then that environment can shut down. It's no longer a cost in the cloud. You deploy the model out to where it lives and then it continues to run with oversight for more of a business persona. Jason, you guys pointed out two use cases in the press conference. One was the health, SaaS health and the other one was the SaaS energy forecasting which you briefly mentioned. Both kind of had this end-to-end kind of life cycle. We've heard the keynote as well, AI life cycle, if you will. But we didn't hear a lot of AI but I want to ask you where does the AI fit into? Because also during the Q and A, you and Brian Harris have responded that you're going to see these vertical manufacturing and other verticals do well with AI, like low-hanging fruit. And he said, with this no ambiguity, you see high returns. I thought that was really interesting. So in IoT, what are the areas that you see that are like high returns from a no ambiguity because it's low-hanging fruit. That makes total sense. I think it's best demonstrated with an example. So we have a customer named First Solar. They create solar panels. One of the leading producers of solar panels in the industry. One of the huge problems that they have is as they're producing these glass panels, there's a lot of high heat transfer and high heat in their manufacturing process. Oftentimes that results in breaking the glass. Now it's hard to identify that and to the point to where they've assigned specific machinery to identify this cracks. But that machine is four inches wide, however the width of the glass is, they will subject a person to following up on that to watch it all the time. Now, it's a binary decision. Is that broken or not? If we can help remove that person out of the process and help expand the evaluation of that glass panel, is it broken or not? And do that at scale, it's a huge savings. And we were talking, that's a high-volume industry. They're running that thing 24-7. If we squeeze an extra point of efficiency out of that utilization, it's huge for them. There's no ambiguity in that. It's either broken or it's not and there's no human involved that's increased cost. That's directly quantifiable. Absolutely, and even the response is binary. Take it off the line, move it into the trash bin. That's something that you can program and repeat and machine learning does a great job of that. A lot of your customers in the Fortune 500 especially are driving new ESG initiatives, sustainability initiatives, becoming a fundamental requirement of suppliers. What's the relationship between IoT and sustainability and what role is SaaS playing in supporting those objectives? Yeah, I think across the board, sustainability continues to be a focus. We're starting to see a crossover between sustainability and environmental concerns. Earlier, we were talking about energy forecasting and the ability to use that directly to help the efficiency of generation and transmission within the utility. So around let's reduce the cost, let's increase uptime, that has absolutely been a focus within that sector. But also as we expand into more of the same weather events that impact energy production, also create catastrophic incidences in coastal areas or with flood. So as a result, SaaS just recently released a solution called floodplain management and it's focused on just that within the six hour window being able to allow these municipalities to talk their constituents about potential flooding areas. The sad fact is the history of this has been they use these 100 year floodplain models and now they're 100 day models, right? These storms are starting to occur which much more frequency and it's critical to be able to have real time on the ground data through the sensors deployed throughout the watershed providing insight into these city managers of what do we do to help our constituents maintain safety? And they're not staffed up to the gills either. They're like, they're running probably some cloud, they're going to be lean and mean, you don't see them with big fat budgets. Absolutely right, and when they do have staffing it's not a bench data scientist, right? That they can pull from at any time. How is the utility a mission energy? You can see a lot of value and sustainability and also just societal benefit. I mean, how you're helping them. What's the state of the art product, workflows, technology that you're deploying out there? For utility, it really comes back to the energy forecasting solution. The big change, some of our oldest clients are in the utility industry and we've serviced them for a long time and it's really been about uptime, maintainability, grid reliability and we've had great success but in all transparency historically it's been a high cost hurdle. There's a massive amount of data that's held in on-premise deployments and the big change is the introduction of the cloud and some of these faster moving engagement models like software as a service that allows us to offer the same benefit that we've been engaged with utilities for 35, 40 years. Now we can do it to small and medium size utilities. And it's also critical infrastructure and just in the stop there, we were talking about IoT edge being in space, you know, is that- The ultimate edge. The ultimate edge is the Earth is round. It's not even edge. It's everywhere. So, you know, satellites is now with space force highly contested, highly constrained environment and congested and contained on out there. A lot of interesting dynamics happening in space. Absolutely and there's a lot of people who's going to be chasing that and what we've done is try to identify a set of partners that we can work with with those particular types of contracts. I saw an interesting spec the other day 47,000 objects currently in space. Assuming they're moving at orbit speed, you can imagine the challenge with tracking and understanding what the exposure and risk is to other objects in space. I saw a story about space junk. How do you do break, fix and space? That's tough. You've got to send in these on but it's getting cheaper and lower cost to put stuff up there. It does and we're seeing more and more with that that those are cheaper objects and end up being the exact junkie reference. What are you seeing in smart cities? I remember several years ago, the previous Boston mayor Marty Walsh, they attended a session and they were talking about trying to make Boston a smart city and the infrastructure's so old. I said, you better start with Tucson maybe where it's nice grids. But has, how has AI affected sort of? I was going to say something but I understand. Smart city development. It was a nightmare. This is never going to happen but it'll happen in pockets. The people won't let it happen. It's not just the infrastructure. Yeah, that's true too. Yeah, inertia. Yeah, you know, oftentimes smart cities is such a high level descriptor that you get lost in what it actually means and we really start to see some compartments that form out of that. There's the transportation, traffic, logistics that's part of it. There's energy use and consumption that's a part of it. And we are starting to see many of these municipalities start to parse it down to something that they can break off and resolve. We have a great example right now where AI is used in Turkey within the area, it's one of the largest cities, densest traffic patterns and they have a huge problem with both managing congestion and being able to manage public transportation around that. So they're starting to use AI principles to identify when there's a potential issue. So Iraq or road construction and then automatically start to reroute the public transport around that. So we are seeing adoption for it. Jason, thanks for sharing all your commentary here on theCUBE, we really appreciate it. I guess my final question would be for you is, actually two part, what's the most important story people should pay attention to coming out of day one here at SAS Explorer? And two, what are you personally most excited about as an industry participant, an engineer who's seen many ways of innovation? First, SAS, some most important story here at the keynote. So I use this story a lot. I came out of industrial engineering school, started a job in proper manufacturing and with all the things you're equipped with, you were just pushing to say, do this, this is a principle you can use and there was a real hesitancy to do it. I had a manager pull me to the side one day and said, you have the worst case of new grad syndrome I've ever encountered and you recognize the pushback it is to do anything different from what we've always done. So if I had any advice is there's options out there for you to improve your business, get started today. And what I'm most excited about is the unique ways that IOT, AI, machine learning can help move that needle. Yeah, and have impact. And have impact, absolutely. And plus the architecture is changing. It's a system architecture. Jason, thanks for coming on theCUBE. Thanks so much for the invite. We'll follow up. It's theCUBE on the ground, on the floor here at SAS Explorer. You got open source, you got streaming analytics, IOT. The game has changed. The AI wave is here, AI enabled infrastructure and software, I'm John Furrier with Dave Vellante. We'll be right back after this short break.