Learn about NexusCon!
âI talk about market transformation forces. I think Volttron has that potential, not necessarily that everyone's going to adopt Volttron, but the fact that this open platform that exists that has some of these functionalities⌠Itâs like a pressure source⌠to move the needle.â
âAndrew Rodgers
Welcome to Nexus, a newsletter and podcast for smart people applying smart building technologyâhosted by James Dice. If youâre new to Nexus, you might want to start here.
The Nexus podcast (Apple | Spotify | YouTube | Other apps) is our chance to explore and learn with the brightest in our industryâtogether. The project is directly funded by listeners like you who have joined the Nexus Pro membership community.
You can join Nexus Pro to get a weekly-ish deep dive, access to the Nexus Vendor Landscape, and invites to exclusive events with a community of smart buildings nerds.
Episode 48 is a conversation with Andrew Rodgers, Co-Founder of ACE IoT.
This is a follow-up from our first episode with Andrew on defining âopen.â
This time we dove into the independent data layer of the stack, and specifically how the open-source software Volttron plays in this arena.
You can find Andrew Rodgers on LinkedIn.
Enjoy!
Music credit: Dream Big by Audiobingerâlicensed under an Attribution-NonCommercial-ShareAlike License.
Note: transcript was created using an imperfect machine learning tool and lightly edited by a human (so you can get the gist). Please forgive errors!
James Dice: [00:00:03] hello friends, welcome to the nexus podcast. I'm your host James dice each week. I fire questions that the leaders of the smart buildings industry to try to figure out where we're headed and how we can get there faster without all the marketing fluff. I'm pushing my learning to the limit. And I'm so glad to have you here following along.
This Episode is a conversation with Andrew Rogers co-founder of ACE IOT. This is a follow-up from our first episode with Andrew on defining open this time we dove into the independent data layer of the stack, and specifically how the open source software Vultron plays in this arena. Â All right, Andrew Rogers. Welcome back to the show. Can you, uh, introduce yourself again for us please?
Andrew Rodgers: [00:00:53] I'm Andrew Rogers and the co-founder of a small company called ACE IOT solutions. We deliver a couple of products and project support around an open source platform called Vultron. Uh, and we have a data Lake infrastructure as a service platform that we offer for facilities and operations technology data.
James Dice: [00:01:12] Absolutely. And you were on episode 20, I believe.
Uh, so it's been six months or so. And we're gonna build on that conversation in this episode, but not re probably not replay a bunch of it. So, uh, if you haven't listened to that episode, you don't necessarily have to go back to it. Uh, but you might, might want to, we'll kind of touch on a lot of the same.
topics, and, and that conversation, we talked about interoperability and openness at each layer of the smart building stack. And it was, I learned a ton. I know that I've been using ever since we had that conversation. Um, so what have you been up to for the last six months?
Andrew Rodgers: [00:01:47] Well, we really had just been focused on building out our platform.
Um, we've, we've been kind of drawn by some of our partners and do a few different. Uh, verticals that are a little unfamiliar to us. So it's been a lot of fun exploring that we've got, um, sort of this converged, you know, obviously our, our sweet spot is in buildings and around buildings, but we're getting drawn into more municipal infrastructure, water supply, that sort of stuff.
So we've been doing some of that. Um, and we've also been working more with utilities on sort of broader, you know, scaled out demand response, and distributed energy resources, which is. Sort of in the, uh, pedigree of Ultron, what it was developed for in the first place, but the market is now kind of catching up to that.
Absolutely.
James Dice: [00:02:30] Yep. And that's what the focus of today is, is on the kind of digging into Vultron and what you guys use it for. But first on the water side, Um, I'll ask you about the hack and Florida, uh, for people that don't know anything about the world outside of buildings, that would be like an industrial use case, uh, or an industrial case study, I guess, is a better term.
W what happened there? Can you kind of fill us in,
Andrew Rodgers: [00:02:55] I mean, I watched that and, and it's so it's so interesting. Um, I have so many thoughts here. There's, there's, you know, A tendency when these things happen to sort of talk about them without a lot of knowledge or like real information about the event.
Um, there were a few like keywords that kept showing up in a lot of, of, uh, news reports. Like there was no firewall. It's like, well, what does that mean? Like, do you mean there was like an internet connection plugged directly into a computer? Probably not. It's probably a router. There was probably a firewall.
Obviously traffic wasn't blocked. Um, and then secondly, like they were running team viewer, you know why they were probably running team viewer. They were probably running team viewer because some draconian IP policy was preventing them from deploying a technology that allowed them to securely manage remote access or, you know, getting the information they needed, uh, you know, data egress, whatever, and like, I always go back to drastic park on this, like life finds a way, right?
Like the more draconian your policy is, and the less you interact with. The folks who are, you know, solving the day-to-day problems and need solutions, the more likely it is, they're going to go around whatever that policy is. And you're going to end up in a situation like this. Should they go around that?
Absolutely not. Right. Like nobody's arguing that it should have happened or that like team viewer should have been running insecured or without the password, you know, being rotated or whatever. Um, but you know, Without better solutions in that space. And if you compare that to what we see, like every day in buildings, ports forwarded into Jason's, that haven't had their passwords changed in 10 years, you know, passwords the same for every building that vendor ever touched.
Like it wasn't the worst thing that could have happened. Um, you know, and then if you really pay attention to what some of the real, like. People who really focus on things like critical infrastructure security, like the other controls that should have been placed were missing as well. Right? Like you don't put in a chemical pump that can pump enough chemical to poison somebody into the water supply.
Right? Like you, there are those kinds of controls that should be in the engineering of a plant like that, that we're missing as well. Um, which I think is the more interesting part, like you know, You can maybe disable whatever security, like there's always some sort of security hole, but like if the pump can't pump more than a certain amount of liquid, like that's a hard limitation that you can use to deliver this, like assuredness that maybe our critical infrastructure deserves.
And finally, the last point is that just obviously as a society, we critically underfund our critical infrastructure. And like, this is just a symptom of that. There will probably be more. Instances of that happening. Um, we're very excited to be able to, to work in that space because we are able to deliver secure data egress for these platforms and enable people to do their jobs better, respond to incidents, you know, um, 24 seven without, needing to open up a team viewer to the SCADA system or, um, you know, drive into the plant.
So. Yeah. What are
James Dice: [00:06:14] some of the use cases that you're deploying as someone that kind of, you know, you didn't originally come from the buildings world, but you guys are primarily doing buildings related work for the most part. So what are some of the, the use cases that you're deploying on the industrial side of things?
Andrew Rodgers: [00:06:29] I mean, it, it very much looks the same. It's connecting, using, you know, some of the legacy protocols that are kind of painful to, to navigate, from a technology perspective and collecting that data, getting it into a normalized form, getting it into a platform where it's easy access for whatever, you know, downstream applications.
So it's, it's more or less the same thing, you know? Simple alarming. you know, I think we kind of got into this a little bit last time about, what is fault detection. So like we can do multi-variate alarms, but I'm always hesitant to talk about doing fault detection, because I think that's like a very, like the people that are really doing that have done a really good job.
And the problem is the. A lot of people say they're doing it. And aren't um, so I don't want to be one of those people who says that's what we're doing when we're not. Um, but we do, you know, support things like, you know, when the chlorine levels high, there's an alarm and somebody gets a text and they can go look at the graphs and see what happened in which pumps are on and that sort of stuff.
So, um, that's, it's the same kind of, you know, all these processes look the same from a conceptual perspective, you've got some product that you're making water. Air conditioned, air, whatever. You've got some critical values that, that contribute to the successful generation of that product. Uh, when they go out of range, you want to know about it and, and you want to be able to see when that happened and what other contributing factors, you know, We're we're causes are playing in that, in that event.
Totally.
James Dice: [00:08:00] That's a good segue into the conversation I want to have today around Vultron. So before we get into Vultron specifically, I include Vultron in like a Nick greater umbrella of. What I call independent data layer, but there are other terms that people use, um, data lakes. I'm sure you have a bunch of other ones as well.
Uh, but let's just, let's just kind of define, like, what is the independent data layer and what sort of makes it unique?
Andrew Rodgers: [00:08:26] Sure. Um, So the independent data layer is really about. You know, if you referencing our, our last talk, I think at the end, you, you forced me to come up with some like one word answer to this question of what is open. And I said choice, and I think that like, that's ultimately the problem that an independent data layer solves for is giving the owner, the operator, the facility manager choice on what kinds of solutions they want to bring to bear for different kinds of problems within the system.
So, you know, it's easy to imagine. There's a lot of like energy management information systems, right? So those can be like independent data layers from doesn't matter who your meter manufacturer is or bringing all that into one MIS and you've got this like independent layer and maybe you have access to that data in some standardized format that you can use other tools, maybe not, but.
That's not a proper EMI S is not necessarily an independent data layer, but it can be. But ideally that independent data later is bringing in data from all kinds of systems. You've got metering data. You've got the facilities, operations data, the actual BAS equipment data, central plant data. Um, you know, battery storage system, data, solar panel, you know, your, your, your PV system data, all of that into one platform where you can start bringing things together.
And now we're talking about things like access control, um, you know, occupancy, you know, we're talking with one, one customer who brings in all of their gym equipment usage into this, this independent data layer where they can manage and make decisions based on, you know, How much equipment is in use at the gym can drive the ventilation level for the gym.
Um, so once you have those kinds of integrations, and then I think there's kind of two very separate classes of independent data layers. There's a lot in the one way data collection side, there are fewer in sort of the two way where you are, are offering not just a normalized interface for the data itself, but then a normalized interface for writing back.
Process variables process set points to those systems. And generally when you do have that two-way control, I think this is kind of maybe muddying the waters between independent data layer and advanced supervisory control, but, you know, Sort of the same as with advanced supervisory control, you kind of keep your low level loops and, um, sort of real time critical stuff out of that system.
And you're just kind of putting in the kind of, overarching set points, um, you know, variables that are, less frequently changing, or just sort of inputs into a real-time process versus Running any of those real-time processes directly in that layer.
James Dice: [00:11:08] Right. Right. And the reason I call it independent is because it's, it's, different than what we have today, which is a full stack point solution, which is kind of like the state-of-the-art and smart buildings right now. So it's basically saying, we're going to chop that up. We're going to, we're going to say, okay, right now we have like an integration layer.
We have a historian database, we have some sort of data model in that story database. And then on top of that, we have applications that provide some sort of, you know, User interface for the user. Right. And so what we're saying with independent data layers, for the most part, we're, we're chopping that in half.
So you're, you're performing the integration, like you said, one or two way, and then you're storing the data all in one place and then a common data model as well. So what about that data model piece? What's sort of the state-of-the-art right now, when it comes to that. Both of us start laughing
Andrew Rodgers: [00:12:00] when we say stayed at dark.
Well, like it's so I don't cry. I think this is the most depressing thing about our industry right now is that this conversation, it just keeps continuing and there doesn't seem to be any actual like resolution and to the point that people are like, Someone, someone responded to a LinkedIn, article I posted recently against standardization, um, like as being like an evil force.
so, so all that to say it's very complicated is, is where things are at right now. Um, in my opinion, we need to standardize. We need like, Everybody in the industry to put their, their, you know, grownup pants on and like agree and find some common ground. Um, we know from experiences in the it world, in technology world, that once you have a common, uh, infrastructure, a common standardization for infrastructure, once you have HTTP, once you have the worldwide web, like there's all of this value, that's unleashed.
And you know, where we're at right now is CompuServe and AOL. you know, we've got literally those walled garden approaches. Can you
James Dice: [00:13:18] explain, can you explain that for those of us that
Andrew Rodgers: [00:13:21] aren't old enough for that? Well, yeah. Yeah. So before the internet, um, As the internet was developing the internet as a, as the internet, like, uh, capital I internet, proper noun, you know, was this thing developed and research and, you know, defense organizations and that's where like all the peer to peer networking that that really makes up like, and I say peer to peer networking, I mean, like routed traffic and IP spend all that stuff was developed.
But, you know, before all of that, there were these. Online service providers. It wasn't really the internet. They were not standardized. They weren't using any of these protocols that were standardized. They were their own walled garden systems, much like the Facebooks of their time. like if you weren't, if you weren't on Facebook, then like you couldn't talk to it.
Aunt Glenda or whoever. Um, you know, so like you were on CompuServe, you're on AOL. You couldn't talk to each other, like, you know, tough crap. So that, that obviously has changed. And it happened once we sort of opened up. and adopted open standards in that infrastructure. you know, otherwise we would all still be like, and I mean, it, the cycle continues, right?
This is not something that's solved by any means. I mean, this is like the iPhone Android. I mean, there's just, this exists everywhere, but it's just so bad in this industry. And everyone recognizes, it's a bad thing. But then people don't seem to be willing to like, do the work to address it. Um, which is why I say it's depressing, but I agree.
James Dice: [00:14:54] Well, I think, I feel like the promise of the independent data layer is that it says we're going to do, you know, one data model, but then like the fine text in the bottom. Basically it says that like, well, it's not going to be anybody else's data model because that's impossible. We don't have one.
And so, so like the subtext is that you're going to be then forcing the data model onto all the applications that are going to be writing on top of it.
Andrew Rodgers: [00:15:19] Right. Th I mean, so, so when you look at the private commercial solutions in this space, they do tend to be very opinionated about the data model. I think that's one of the things that is attractive about Vultron and, and how we've built our platform, you know, our platform and, and there'll be a little tricky today talking, separating some of the value we've added on top of Vultron in our platform versus what's built into Vultron itself.
I apologize. I will try to be clear about that. Yeah, Vultron in itself is not opinionated at all. It treats everything sort of natively as it finds it in the system that's being integrated. Um, and you know, for backnet, for instance, that just means you get back net device IDs, and, and point addresses and, and registers.
Right. Um,
James Dice: [00:16:07] Can you back up and, and let's, let's do dive into Vultron. Now we talked a little bit about it on the last episode, but assuming that people didn't hear that, what is
Andrew Rodgers: [00:16:16] sorry. Yeah. Yeah. Um, so, so Vultron is a open source technology project. It was funded by DOE led by Pacific Northwest national laboratory.
Really around the idea that DOE was doing all these really interesting. What they now call grid interactive efficient building projects, Gabs. Um, at the time I don't think it was that term was really a term of art. Um, I
James Dice: [00:16:42] think it was grid integrated buildings back then
Andrew Rodgers: [00:16:45] district. I like better sources, et cetera.
Um, and. They had this problem, which I've done a decent amount of work in research, uh, research computation, not, not in the national labs like James has, but in the more in the academic side, um, there's a lot of just wasted effort in like recreating. Simple boiler plate stuff that you need. Like, you know, for instance, if you're doing research on the absolutely best, optimization strategy, like every grant would have funding in it where you do the integration again.
So like, you know, everybody's like connecting the test, you know, uh, air handler to mat lab. In every single proposal, like it's the same work being done by different grad students. And like, you don't get, you don't even get the same grad student who's already done it once before. It's a new grad student who has to learn the whole thing again.
So DOE was really trying to say, like, let's build, let's build some infrastructure that enables this research to work faster and cheaper. So that's really where Vultron came from was, enabling. Rapid testing, safe, uh, operations. So like when you use Vultron as a two way platform, you know, it has all the semantics built into it for managing what priority things get written to in the backnet control or those sorts of things.
It has automatic rollback. So your application, if your application falls over Vultron will detect that and roll back to whatever the previous set point was before your application overrode it. Um, so it's got a lot of really cool features around that in the building space. And then it was more broadly, you know, not just the building systems, but designed to sort of bridge that gap between.
Uh, building automation systems back net mod bust, that sort of stuff that was in the building envelope and the grid scale, you know, utility distribution system protocols, like the MP3 and open ADR and, and those sorts of things. So it was really. Bill as an infrastructure platform that allows people to build applications rapidly and cost-effectively, um, by kind of solving some of these boilerplate problems that you just solve over and over otherwise.
Yeah.
James Dice: [00:19:03] And so how do you, you, you basically took it, got turned into an open source project through the eclipse foundation. I think that's been yep. And then people like you, there's probably other firms like you guys then build, build onto it and create your own product. Is that how it works?
Andrew Rodgers: [00:19:18] That's right.
Yeah. And so like for us, You know, our cloud platform is very much our product. There's not a lot. I mean, Vultron is involved in the data aggregation and collection, but then once it hits our platform, like the data Lake and the visualization and all that stuff is completely different technologies that we've assembled and put together into that platform.
Um, but we, we try very hard to keep the edge, uh, resources, pretty much vanilla Vultron. just so that. When one of our customers wants to deploy some sort of Vultron application, that's not ours, they have that capability. and that, that choice, again, back to, you know, kind of built into our ethos is giving our customers as much choice as we possibly can while maintaining, I mean, the, the risk here or the, the cost model, how you think about this as a business problem really is, you know, the more choice, the more risk.
And you have to pay for that risk somehow. So it's always, for us, it's always walking this really, you know, trying to find the best balance point for our we're giving our customers as much choice as possible while still like keeping it manageable, where we can offer a good product for a good price.
James Dice: [00:20:29] Totally.
Hey guys, just another quick note from our sponsor nexus labs. And then we'll get back to the show. This episode is brought to you by nexus foundations, our introductory course on the smart buildings industry. If you're new to the industry, this course is for you. If you're an industry vet, but want to understand how technology is changing things.
This course is also for you. The alumni are raving about the content, which they say pulls it all together, and they also love getting to meet the other students on the weekly zoom calls and in the private chat room, you can find out more about the course@courses.nexus lab. Start online. All right, back to the interview.
Well, the reason I wanted to dive into this to stay with you is I feel like this. Advanced supervisory control piece is hot, right? And the independent data layer is hot and they'll try and you know, is a heat tool for both of those things. Right. And. I wanted to ask you around, like, why would someone build their own piece, like proprietary black box inside the building to do that two way integration and why shouldn't they, I guess, and you just kind of, you said it, but why don't you say it like to the, to everyone else?
Uh, why shouldn't we
Andrew Rodgers: [00:21:38] do that? Yeah, I mean,
I would say there are good reasons to do that. And there are good reasons to not do that. I wouldn't say that, like, you know, if your business model is a traditional business model in this space of capturing a customer and like, Tying them down as hard as you can and giving them as little choice as possible, then like, yeah, write your own.
Right. Like use the most obscure stuff you can because that's how you're going to like, keep, keep them locked. Um, but for her calls it, lock them and lose them. Yeah. Yeah. Yeah. He does. Uh, so for the rest of us though, like who are trying to maybe see a better future in this space, you know, I would say this.
And I, I do say this to the Vultron team at PNNL, um, I'm, you know, one of the most active community members outside of the lab system, for sure. I'm also one of the ones that's like first to question every, but you know, there are things in the Vultron project that I don't agree with. Like, I wouldn't have done them that way.
I don't like them. I don't, you know, I don't think it's the right way to do something, but. The value of being in an ecosystem where there are other people using that technology, um, where we have sort of this plausible, presentation to our customers that like, yes, we're a small company, but we're using this technology that is open and, you know, there are other providers for, so if you decide that, like we're not.
Delivering you can fire us and you're not left holding a bag of some proprietary crap that you can't do anything with. And that value, you know, again, it's a very different perspective, right? Like, of course you want to tell your customers that if they fire you, they're going to be left. You know, they're going to have huge amounts of costs to recreate your so, you know, that's why I say like, From my perspective.
And I think maybe from the reality you would like to see in the world, no, no one should be doing it that way, but there, if that's your business model, that's your business model. And I mean, I'm not here to, I'm probably not going to convince you otherwise. And, and since that super as this model.
Yeah. So, um, but I would encourage you to investigate maybe like maybe we should be working on delivering value in a little more, um, Again, interoperable way. Like our goal is we solve the data aggregation, putting the interfaces together to allow control this advanced, you know, the advanced supervisory control.
We solve that problem really well. We don't compete with. Fault detection companies like we would, we love it. When our customers take our platform, bring the data back to a common layer and then hire a fault detection company to make sense of that data or deliver some, you know, direct value. And then turn around and say, Oh, well, our AFTD company, like they do everything really well, except for this one weird system that like, we're the only ones that have, because, you know, we made dumb decisions in 1980 or whatever.
Right. And then they, but then because they have this independent data layer, they have the choice to say, Oh, well, we're going to invest, you know, a few months of development time with our team to build something special that does that for us. And we have that flexibility. Whereas if they had just went with.
You know, an AFTD vendors, data collection tool that kept everything in a closed ecosystem and wasn't open and accessible. They would not be able to unlock that value. So that's where I think the independent data layer really starts to, you know, start to shine. And who
James Dice: [00:25:12] are the other people that are in the community, the Vultron community, besides the PNNL folks who, who is the people contributing
Andrew Rodgers: [00:25:19] back to them?
So there's a core user group. It pretty much each national lab that, that obviously, um, that is still using Vultron and working with Vultron, um, companies offering Vultron as a service, There's not that many of them, I would struggle to, to name one, who's like doing it in a way that's not, um, competing with any of the existing players.
Like there are companies that do fault detection and then they use Vultron under the hood to do that. There are companies who are doing, you know, completely managed building optimization services using Vultron. Um, but. From, we are probably some of the only ones who are out there, like selling it purely as the platform play.
Um, yeah.
James Dice: [00:26:08] And how do you guys sell it when it's just, it's not your product added on top. It's an open source piece of software that you're setting up. How does that
Andrew Rodgers: [00:26:15] work? The way I like to look at it is, you know, Anybody can go download Vultron off, get hub, start running it on their computer, run it, you know, install it on a raspberry PI if that's what they want to do.
Um, anybody can do that. What we bring is tools that allow us to deploy faster, collect data, better, more efficiently. Uh, monitor for changes in the network, monitor the hardware platforms we deploy, um, and maintain the health of those hardware platforms. We're really selling a service around reliable data acquisition.
That Vultron is a plays a central role in, but it's really the suite of, of, you know, the monitoring main, maintenance, all that stuff that we're offering on top of Vultron is really where our value comes from. I see.
James Dice: [00:27:10] Let's talk about the grid interaction piece of it a little bit. So Vultron has that, I mean, that was one of the intentions when it was created, it was just to be used for a building to interact with the utility.
Right. Um, what's the status of that today? Uh, and what are you seeing on, on real projects? Because I think where I'm at is. You know, leaving Enrail six months, seven, eight months ago, however long that was, I'm kind of out of that world, but like there's still this like opinion of buying where it's like, we're talking about a lot of hype for this and well-reasoned hype.
Like we need to decarbonize the grid in order to do that. We need to have load flexibility totally. 100% behind that. But what we haven't talked about is like, What buildings are out there doing it and why are they doing it? And does a building owner care about interacting with the grid? And so I guess what I'm asking you is like, give me an update on like load flexibility in grid interaction and 20, 21, like where are we at?
Andrew Rodgers: [00:28:12] Yeah. So, I mean, Vultron has been being used for, for. Pilot demonstration projects with utilities in this space since its inception. I mean, I, you know, probably day one with Vultron was that was the actual use case for it. Um, what we see is utilities starting to talk to us and say, Hey, we've been working with, you know, Oak Ridge national lab, Pacific Northwest national lab for years now on these pilots.
Can we get this thing to scale? Like, is this ready now? Can we actually do this? Can we start seeing aggregate benefit that is going to justify the investment? Um, at the same time, I think you're seeing things happen in buildings that are, you know, in the, the actual owner market and buildings that are pushing, you know, local law, 97 in New York, right?
Like, It doesn't directly affect grid interactivity, but because it's making people go out and touch these buildings and in P you know, do data collection, do, a lot deeper analysis of how energy is being used, then you start to see the movement that allows, you know, demand response or, you know, active load management.
To be kind of shoehorned into existing projects. Like, no, I think, I think we're, I think there's a, there's a huge shift, but I still think in 2021, no one is going to go out and say, Oh, it is worth it for us to go deploy all new infrastructure in our portfolio to achieve a load flexibility. Like nobody's going to do that.
But when it's, Hey, We're going to install vehicle chargers because our tenants are demanding electric vehicle infrastructure to install electric vehicle infrastructure. We've got to put 200 KW KWH a batteries in because that's the only way we can do it without completely upgrading our utility service, which is going to cost a million dollars.
That's when you start seeing, Oh, well for, you know, an extra tens of thousands of dollars, we can use this for demand flexibility. We can do our peak shaving. Like we'd get all these other ancillary benefits because there's already capital moving. I think that's where we'll see the most adoption on the building side.
The utilities are actually really interested in this stuff. On the residential side as well. So we're working with a few utilities on residential demonstration projects where you start asking the question, okay. People are putting, you know, Tesla power walls out there and rooftop, solar and electric vehicle in the garage and a water heater that's connected and you know, a smart thermostat that's, that's enabling control of the HVC.
Like if we start aggregating a hundred houses, a thousand houses, 10,000 houses, like. We can actually start making a dent in peak demand. Um, so I think, I think we're approaching market readiness. I wouldn't say we're like at full market readiness yet, but I think the market is in a much different place than it was two or three years ago.
Um,
James Dice: [00:31:22] and so w what does the stack look like from a Voltron perspective with that aggregation? You got a local node in each of the houses or in each of the buildings, and then you have
Andrew Rodgers: [00:31:31] one aggregator side. We've done. We've done some demonstration projects where, um, you know, Vultron is just operating in the cloud, kind of being an integration layer between a bunch of different cloud API APIs.
But when you start talking, especially in rural, America where things like battery storage can make a huge difference in electricity, reliability on long distribution lines, right? Like you put a few. Tens of KWS of our KWH is a batteries out there. And, you know, a storm falls a tree across the line somewhere and you can keep people's like critical services up like that can have huge impact.
I mean, Texas, Texas, this winter, I think is a great example of, Such a big failure at a massive scale that like the communications infrastructure failed too. So like, even if you had smart thermostats, it didn't matter because if they're going through a cloud round trip, they weren't getting out.
Um, right. So, so there's this desire on the utility side to really explore more resilient applications, right? So like if you've got a battery attached to a house and you've got solar panels on the roof, Can you Island that house and, you know, shut off all the non-critical loads and keep the refrigerator and the temperature in the house above freezing for, you know, 48 hours.
Like, and for that you need edge compute capacity. You need edge integrations to, to really do deliver that in a way that's fault tolerant. Totally.
James Dice: [00:33:05] Very cool. So let's talk a little bit about, Vultron sort of in context to the other ways of doing things. So we, we, we talked about the fully open model.
What are the other ways that like the independent data layer is being provided today? Like one of them is that I have, I'm curious about, you've talked to Mike Berman about SmartCore. Um, I'm sure. Like how, how does ultra and compare
it
Andrew Rodgers: [00:33:28] to smart core? So Mike and I had a really great conversation.
I think both of us are excited to work together. It's just finding the right application. Um, but smart core is much more focused on. Like day to day operations of the building and interaction of, um, sort of, yeah. User interfaces in Vultron has almost zero focus on that. Right. Vultron is about backend services, enabling, you know, these complicated control algorithms that need data from a lot of different sources and need to right.
Set points back out to a lot of different systems. Um, And I think the other thing smart core does not focus on at all is that grid integration piece. So Mike and I had talked about some applications where smart core might be doing all of the systems, uh, dashboarding and, and, you know, HMI, all that stuff in a building.
And then there would be a Vultron instance that it would communicate with as well in order to manage the grid, to building an interaction. Um, so I think, I think. Specifically with SmartCore. I don't think Vultron has, uh, Vultron is just a very different approach. Um, and maybe you can get to the same place, both ways, but I think depending on who you are, where you're coming from, you might would pick one of those tools versus the other versus necessarily like the specific problem you're trying to solve.
Other than like the UI stuff. That's all smart core. And, and H and Vultron just really, doesn't doesn't even try to address that. Yeah.
James Dice: [00:35:02] Got it. What about the other alternatives, I guess, doing sort of the open independent data layer, uh, on Voltron?
Andrew Rodgers: [00:35:10] Well,
I don't know that there are a lot of, there's not a lot of alternatives for a truly open, independent data layer right now. Like we see a lot of vendors doing their own, Models their own sort of proprietary solutions. And the question I'm always left with kind of, or, or the question I'm always asking when I review those systems is, are like, Where is the choice?
Like what happens if I want to separate myself from this company in the future? Where, where does that leave me? Um, you know, do I have a path where I can, you know, take my data and reimplement, another set of infrastructure that allow me to collect the data I need? Like, does that mean I need to completely redo all of my, my applications.
so that's. Where I think Vultron has sort of a unique focus. Um, there doesn't seem to be a lot of market pressure otherwise for these companies to really be open. Um, the other thing that you know,
it seems like there's a lot of effort on the companies who are saying, like, we're delivering an API for your building, those kinds of value propositions. Yeah. Yeah, they're controlling all of that and you know, if, if you're just using the model they have, you don't have control or choice in, you know, if you want to add additional features to that model that are specific to your use case, It's not clear to me how someone can, can deploy something that implements a, you know, somewhat proprietary model and decide you wanted to swap that with something else.
It's not clear how you would go about recreating that. Um, you know, data collection infrastructure in a way that, that, you know, was self-actualized
James Dice: [00:36:56] absolutely. Especially when you start to talk about connecting a bunch of different applications to it. So if you just have, you know, if you've pictured just a point solution today, you have one application connected to the data layer.
Right? That's great. It could be pretty easy to swap out components there, but when you start to talk about like the true value proposition of the independent data layer, when you have, you know, all these different applications plugged in, right. Uh, and now you're swapping out the infrastructure piece that, yeah, that does sound very, very difficult.
Um, interesting. So how about on the, let's go back to Vultron a little bit the advanced supervisory control today, like. I think I'm always wanting updates from the field because, um, it's still an area where I think, I think you have about 50, 50 of like there are a ton of startups that obviously believe that it's the future.
Right. I think on my vendor landscape, there's, it's up to like 15 to 20 now, um, of people doing this, Cloud application providing advanced supervisory control. And then there's the other 50% of the people that it's like never going to happen. Right? Like this is absurd, right. That we would provide control from somewhere else besides the automation system or the system that was designed to control the systems.
So. Well, like, can you provide an update on, on the, kind of where the market's at today? Um, I think this is an important question because in order to make things like grid flexibility happen, right? Like we talk about, Oh, we need grid flexibility, but in order to make it happen, we as an industry have to kind of get on board with the advanced supervisory control.
It's just the fact.
Andrew Rodgers: [00:38:40] Yeah. And, and I mean, I. I guess I just have commitment issues in all of these questions. Like you'll ask, well, is it this way or this way? And I'm always like, well, we just decided and just support both because it turns out like, you know, that's, that's where we, so, so when we talk about data modeling, like we are not opinionated about data modeling, we built out infrastructure to support whatever models our customers bring to us.
We talk about. You know, edge control versus cloud control. We have customers who were saying, Hey, we are actually already have an independent data layer. You know, we did it early. We were very advanced. Uh, but we didn't get to way control and like our vendor's not going to do that.
Or the system we built out internally is not going to do that. Can you go deploy Vultron and just give us an API for riding back, like these key set points from your cloud. Like that's, that's like super easy for us to do, but then if you want to like drink the Vultron Kool-Aid maybe you say, Hey, can you put Vultron in our buildings?
And give us a platform for us to deploy these applications that we want to develop that are doing this automatically and run at the edge. And the answer there is yes, as well. Like that's, that's why we kind of focus on doing this, um, you know, sticking to as close to vanilla Vultron as we can.
And those edge applications is so that we can take these packaged, you know, what vault in Vultron world as an agent, um, which isn't, you know, basically an independent application that runs in the platform context with access to the data that you provide it. Yeah. Via sort of an access control list. Um, so if you're familiar with deploying things into cloud environments, You know, you can choose very granularly.
What data, a particular application has access to. And there are open source. They're really great open source things that the labs have built that are available as Vultron agents. There's, um, there's one called intelligent load control that PNNL has built out. That's all around sort of doing peak demand, limiting with comfort in mind.
So you can kind of prioritize spaces, um, in your building and then set a target, load profile that you want your building to maintain. And it will, kind of. Shed load in a weighted way that tries to keep your priorities zones happy. Um, and then maybe we'll give up on shedding load in order to maintain a certain like, threshold of, of comfort that you're not willing to go outside of.
Um, and so that's like a Vultron agent that you can, as again, open source, you can download it, you can install it, run it on your building and achieve, you know, peak load reduction. Pretty pretty straightforward. so that's like the fully drinking, the Vultron Kool-Aid way to do it. It's running in the edge.
It's resilient. It's kind of giving you all the value proposition that people would say that your existing BAS gives you. But it's still a supervisory system that can pull in data from a lot of different sources instead of just the BAS itself. And you get cloud or, you know, external connectivity oversight.
So you can easily connect, bring all that data back to one place while the actual decision-making is happening at the edge. Um, one of the things, and I guess this is my experience in the industrial world, but I was amazed when I started. In building automation and found out that a lot of these startups you're talking about who do say the independent data layer, like a lot of it it's like they install a VPN and then they have like a backnet client in the cloud, somewhere talking to the BAS over this VPN link.
And if the connectivity goes down, you don't get data. Like there's all kinds of reasons not to do that. and it just in an industrial world that would have never flown because You don't want to give up having your data just because your internet connections down. Um, you proxy that somewhere locally Vultron does all that for you.
So not only can your edge device be sitting there running some optimization strategy, it's also collecting the data, buffering it locally, forwarding it to the cloud. So you still have sort of a portfolio wide supervisory view of what's going on, but you can actually deploy those control. Algorithms directly to the edge device.
Absolutely.
James Dice: [00:43:07] Interesting. That that's a good enough for me like that you had me at like our clients are asking for this, like, like I feel like we just need more case studies around actual building owners investing in these types of controls, because I still think. I again, I already said it. I still think that we're, we're not quite as an industry decided that this is a good idea at this point.
Yeah. Yeah. And I've heard several people that like several building owners, large building owners that have, I just heard about recently that are investing in this approach and we just don't have enough public case studies on it yet.
Andrew Rodgers: [00:43:45] Yeah. I mean, it's, One of those challenges, because if you're really good at it, it's definitely a competitive advantage.
Um, and you know, if you're a big either portfolio, owner or manager, you don't necessarily want to talk and give away all of this, you know, great ways that you're improving your return on your investment. Um, so it's, it's a little tricky. I, you know, I still think, you know, Joe Gaspar, Tony always makes this point that like, CRE is the hardest place to do this.
And you know, Â REITs are really hard to do this in because everything is driven by this value proposition and return on investment. Um, but I do think from what I've seen is like some of the most interesting things are being done by the private, um, private and commercial owners instead of the, the campuses and universities and such.
I think, I think, I think it's that three 3,300 thing. Um, that drives that, you know, Â and then like the future's already here. It's just not evenly distributed. Like somewhere between this to stay between those two cliches is why they're a few property owners doing really cool stuff today. And we don't hear about it a lot. Right, right.
James Dice: [00:45:05] Totally. Well, this has been fun.
I think that's, that's the end of my current round of questioning for you. Did you have anything else you wanted to kind of get off your chest? This,
Andrew Rodgers: [00:45:15] this round? Uh, I don't know. Um, this was, this is always good to catch up. I appreciate you having me for sure. Um, you know, I'm always happy to talk to anyone about Vultron.
it's, it's very exciting to me as a technology. Um, I've just, I I've talked about, I think you'll, you'll see, like I post things on LinkedIn a lot and I talk about market transformation forces. I think Vultron has that potential, not necessarily that everyone's going to adopt Vultron, but that, that, open platform that exists that does some of these functionalities is just, uh, a pressure point.
Like it's our pressure source, I should say, to move the needle and. you know, that may mean that like know, Mike Ruman at SmartCore has, you know, CS. Voltron's like, Hey, if they can do it, I can do it. Um, it might mean that, you know, someone has to have to bid against a Vultron solution and it encourages them to change something about their product that makes it more open.
Like there's all kinds of ways. you know, The existence of something like that can change a market. Um, I think I posted recently something, I was really amazed, like this really great industrial controller, based on the Arduino platform, very low cost, really cool features, and to see how the accessibility of embedded development of industrial control using open technologies.
All of that's been moved by like some crazy professors in Italy who wanted to make embedded development easier for their staff. It's like Arduino as a market force is a really interesting story. I think it has a lot of parallels in what something like Vultron and smart SmartCore can do in this sector.
James Dice: [00:47:06] Amazing well to keep, keep getting updated. Uh, I always love your LinkedIn posts and please keep it up and we'll have to talk more about, uh, uh, Andrew on interoperability. That's what I'm going to call it. Uh, we'll have to talk about it after we, after we hit stuff here. So thanks for coming on the show.
Thank
Andrew Rodgers: [00:47:24] you.
James Dice: [00:47:25]
All right, friends. Thanks for listening to this episode of the nexus podcast for more episodes like this, and to get the weekly nexus newsletter, which by the way, readers have said is the best way to stay up to date on the future of the smart building industry. Please subscribe@nexuslabs.online. You can find the show notes for this conversation there as well. Have a great day.
âI talk about market transformation forces. I think Volttron has that potential, not necessarily that everyone's going to adopt Volttron, but the fact that this open platform that exists that has some of these functionalities⌠Itâs like a pressure source⌠to move the needle.â
âAndrew Rodgers
Welcome to Nexus, a newsletter and podcast for smart people applying smart building technologyâhosted by James Dice. If youâre new to Nexus, you might want to start here.
The Nexus podcast (Apple | Spotify | YouTube | Other apps) is our chance to explore and learn with the brightest in our industryâtogether. The project is directly funded by listeners like you who have joined the Nexus Pro membership community.
You can join Nexus Pro to get a weekly-ish deep dive, access to the Nexus Vendor Landscape, and invites to exclusive events with a community of smart buildings nerds.
Episode 48 is a conversation with Andrew Rodgers, Co-Founder of ACE IoT.
This is a follow-up from our first episode with Andrew on defining âopen.â
This time we dove into the independent data layer of the stack, and specifically how the open-source software Volttron plays in this arena.
You can find Andrew Rodgers on LinkedIn.
Enjoy!
Music credit: Dream Big by Audiobingerâlicensed under an Attribution-NonCommercial-ShareAlike License.
Note: transcript was created using an imperfect machine learning tool and lightly edited by a human (so you can get the gist). Please forgive errors!
James Dice: [00:00:03] hello friends, welcome to the nexus podcast. I'm your host James dice each week. I fire questions that the leaders of the smart buildings industry to try to figure out where we're headed and how we can get there faster without all the marketing fluff. I'm pushing my learning to the limit. And I'm so glad to have you here following along.
This Episode is a conversation with Andrew Rogers co-founder of ACE IOT. This is a follow-up from our first episode with Andrew on defining open this time we dove into the independent data layer of the stack, and specifically how the open source software Vultron plays in this arena. Â All right, Andrew Rogers. Welcome back to the show. Can you, uh, introduce yourself again for us please?
Andrew Rodgers: [00:00:53] I'm Andrew Rogers and the co-founder of a small company called ACE IOT solutions. We deliver a couple of products and project support around an open source platform called Vultron. Uh, and we have a data Lake infrastructure as a service platform that we offer for facilities and operations technology data.
James Dice: [00:01:12] Absolutely. And you were on episode 20, I believe.
Uh, so it's been six months or so. And we're gonna build on that conversation in this episode, but not re probably not replay a bunch of it. So, uh, if you haven't listened to that episode, you don't necessarily have to go back to it. Uh, but you might, might want to, we'll kind of touch on a lot of the same.
topics, and, and that conversation, we talked about interoperability and openness at each layer of the smart building stack. And it was, I learned a ton. I know that I've been using ever since we had that conversation. Um, so what have you been up to for the last six months?
Andrew Rodgers: [00:01:47] Well, we really had just been focused on building out our platform.
Um, we've, we've been kind of drawn by some of our partners and do a few different. Uh, verticals that are a little unfamiliar to us. So it's been a lot of fun exploring that we've got, um, sort of this converged, you know, obviously our, our sweet spot is in buildings and around buildings, but we're getting drawn into more municipal infrastructure, water supply, that sort of stuff.
So we've been doing some of that. Um, and we've also been working more with utilities on sort of broader, you know, scaled out demand response, and distributed energy resources, which is. Sort of in the, uh, pedigree of Ultron, what it was developed for in the first place, but the market is now kind of catching up to that.
Absolutely.
James Dice: [00:02:30] Yep. And that's what the focus of today is, is on the kind of digging into Vultron and what you guys use it for. But first on the water side, Um, I'll ask you about the hack and Florida, uh, for people that don't know anything about the world outside of buildings, that would be like an industrial use case, uh, or an industrial case study, I guess, is a better term.
W what happened there? Can you kind of fill us in,
Andrew Rodgers: [00:02:55] I mean, I watched that and, and it's so it's so interesting. Um, I have so many thoughts here. There's, there's, you know, A tendency when these things happen to sort of talk about them without a lot of knowledge or like real information about the event.
Um, there were a few like keywords that kept showing up in a lot of, of, uh, news reports. Like there was no firewall. It's like, well, what does that mean? Like, do you mean there was like an internet connection plugged directly into a computer? Probably not. It's probably a router. There was probably a firewall.
Obviously traffic wasn't blocked. Um, and then secondly, like they were running team viewer, you know why they were probably running team viewer. They were probably running team viewer because some draconian IP policy was preventing them from deploying a technology that allowed them to securely manage remote access or, you know, getting the information they needed, uh, you know, data egress, whatever, and like, I always go back to drastic park on this, like life finds a way, right?
Like the more draconian your policy is, and the less you interact with. The folks who are, you know, solving the day-to-day problems and need solutions, the more likely it is, they're going to go around whatever that policy is. And you're going to end up in a situation like this. Should they go around that?
Absolutely not. Right. Like nobody's arguing that it should have happened or that like team viewer should have been running insecured or without the password, you know, being rotated or whatever. Um, but you know, Without better solutions in that space. And if you compare that to what we see, like every day in buildings, ports forwarded into Jason's, that haven't had their passwords changed in 10 years, you know, passwords the same for every building that vendor ever touched.
Like it wasn't the worst thing that could have happened. Um, you know, and then if you really pay attention to what some of the real, like. People who really focus on things like critical infrastructure security, like the other controls that should have been placed were missing as well. Right? Like you don't put in a chemical pump that can pump enough chemical to poison somebody into the water supply.
Right? Like you, there are those kinds of controls that should be in the engineering of a plant like that, that we're missing as well. Um, which I think is the more interesting part, like you know, You can maybe disable whatever security, like there's always some sort of security hole, but like if the pump can't pump more than a certain amount of liquid, like that's a hard limitation that you can use to deliver this, like assuredness that maybe our critical infrastructure deserves.
And finally, the last point is that just obviously as a society, we critically underfund our critical infrastructure. And like, this is just a symptom of that. There will probably be more. Instances of that happening. Um, we're very excited to be able to, to work in that space because we are able to deliver secure data egress for these platforms and enable people to do their jobs better, respond to incidents, you know, um, 24 seven without, needing to open up a team viewer to the SCADA system or, um, you know, drive into the plant.
So. Yeah. What are
James Dice: [00:06:14] some of the use cases that you're deploying as someone that kind of, you know, you didn't originally come from the buildings world, but you guys are primarily doing buildings related work for the most part. So what are some of the, the use cases that you're deploying on the industrial side of things?
Andrew Rodgers: [00:06:29] I mean, it, it very much looks the same. It's connecting, using, you know, some of the legacy protocols that are kind of painful to, to navigate, from a technology perspective and collecting that data, getting it into a normalized form, getting it into a platform where it's easy access for whatever, you know, downstream applications.
So it's, it's more or less the same thing, you know? Simple alarming. you know, I think we kind of got into this a little bit last time about, what is fault detection. So like we can do multi-variate alarms, but I'm always hesitant to talk about doing fault detection, because I think that's like a very, like the people that are really doing that have done a really good job.
And the problem is the. A lot of people say they're doing it. And aren't um, so I don't want to be one of those people who says that's what we're doing when we're not. Um, but we do, you know, support things like, you know, when the chlorine levels high, there's an alarm and somebody gets a text and they can go look at the graphs and see what happened in which pumps are on and that sort of stuff.
So, um, that's, it's the same kind of, you know, all these processes look the same from a conceptual perspective, you've got some product that you're making water. Air conditioned, air, whatever. You've got some critical values that, that contribute to the successful generation of that product. Uh, when they go out of range, you want to know about it and, and you want to be able to see when that happened and what other contributing factors, you know, We're we're causes are playing in that, in that event.
Totally.
James Dice: [00:08:00] That's a good segue into the conversation I want to have today around Vultron. So before we get into Vultron specifically, I include Vultron in like a Nick greater umbrella of. What I call independent data layer, but there are other terms that people use, um, data lakes. I'm sure you have a bunch of other ones as well.
Uh, but let's just, let's just kind of define, like, what is the independent data layer and what sort of makes it unique?
Andrew Rodgers: [00:08:26] Sure. Um, So the independent data layer is really about. You know, if you referencing our, our last talk, I think at the end, you, you forced me to come up with some like one word answer to this question of what is open. And I said choice, and I think that like, that's ultimately the problem that an independent data layer solves for is giving the owner, the operator, the facility manager choice on what kinds of solutions they want to bring to bear for different kinds of problems within the system.
So, you know, it's easy to imagine. There's a lot of like energy management information systems, right? So those can be like independent data layers from doesn't matter who your meter manufacturer is or bringing all that into one MIS and you've got this like independent layer and maybe you have access to that data in some standardized format that you can use other tools, maybe not, but.
That's not a proper EMI S is not necessarily an independent data layer, but it can be. But ideally that independent data later is bringing in data from all kinds of systems. You've got metering data. You've got the facilities, operations data, the actual BAS equipment data, central plant data. Um, you know, battery storage system, data, solar panel, you know, your, your, your PV system data, all of that into one platform where you can start bringing things together.
And now we're talking about things like access control, um, you know, occupancy, you know, we're talking with one, one customer who brings in all of their gym equipment usage into this, this independent data layer where they can manage and make decisions based on, you know, How much equipment is in use at the gym can drive the ventilation level for the gym.
Um, so once you have those kinds of integrations, and then I think there's kind of two very separate classes of independent data layers. There's a lot in the one way data collection side, there are fewer in sort of the two way where you are, are offering not just a normalized interface for the data itself, but then a normalized interface for writing back.
Process variables process set points to those systems. And generally when you do have that two-way control, I think this is kind of maybe muddying the waters between independent data layer and advanced supervisory control, but, you know, Sort of the same as with advanced supervisory control, you kind of keep your low level loops and, um, sort of real time critical stuff out of that system.
And you're just kind of putting in the kind of, overarching set points, um, you know, variables that are, less frequently changing, or just sort of inputs into a real-time process versus Running any of those real-time processes directly in that layer.
James Dice: [00:11:08] Right. Right. And the reason I call it independent is because it's, it's, different than what we have today, which is a full stack point solution, which is kind of like the state-of-the-art and smart buildings right now. So it's basically saying, we're going to chop that up. We're going to, we're going to say, okay, right now we have like an integration layer.
We have a historian database, we have some sort of data model in that story database. And then on top of that, we have applications that provide some sort of, you know, User interface for the user. Right. And so what we're saying with independent data layers, for the most part, we're, we're chopping that in half.
So you're, you're performing the integration, like you said, one or two way, and then you're storing the data all in one place and then a common data model as well. So what about that data model piece? What's sort of the state-of-the-art right now, when it comes to that. Both of us start laughing
Andrew Rodgers: [00:12:00] when we say stayed at dark.
Well, like it's so I don't cry. I think this is the most depressing thing about our industry right now is that this conversation, it just keeps continuing and there doesn't seem to be any actual like resolution and to the point that people are like, Someone, someone responded to a LinkedIn, article I posted recently against standardization, um, like as being like an evil force.
so, so all that to say it's very complicated is, is where things are at right now. Um, in my opinion, we need to standardize. We need like, Everybody in the industry to put their, their, you know, grownup pants on and like agree and find some common ground. Um, we know from experiences in the it world, in technology world, that once you have a common, uh, infrastructure, a common standardization for infrastructure, once you have HTTP, once you have the worldwide web, like there's all of this value, that's unleashed.
And you know, where we're at right now is CompuServe and AOL. you know, we've got literally those walled garden approaches. Can you
James Dice: [00:13:18] explain, can you explain that for those of us that
Andrew Rodgers: [00:13:21] aren't old enough for that? Well, yeah. Yeah. So before the internet, um, As the internet was developing the internet as a, as the internet, like, uh, capital I internet, proper noun, you know, was this thing developed and research and, you know, defense organizations and that's where like all the peer to peer networking that that really makes up like, and I say peer to peer networking, I mean, like routed traffic and IP spend all that stuff was developed.
But, you know, before all of that, there were these. Online service providers. It wasn't really the internet. They were not standardized. They weren't using any of these protocols that were standardized. They were their own walled garden systems, much like the Facebooks of their time. like if you weren't, if you weren't on Facebook, then like you couldn't talk to it.
Aunt Glenda or whoever. Um, you know, so like you were on CompuServe, you're on AOL. You couldn't talk to each other, like, you know, tough crap. So that, that obviously has changed. And it happened once we sort of opened up. and adopted open standards in that infrastructure. you know, otherwise we would all still be like, and I mean, it, the cycle continues, right?
This is not something that's solved by any means. I mean, this is like the iPhone Android. I mean, there's just, this exists everywhere, but it's just so bad in this industry. And everyone recognizes, it's a bad thing. But then people don't seem to be willing to like, do the work to address it. Um, which is why I say it's depressing, but I agree.
James Dice: [00:14:54] Well, I think, I feel like the promise of the independent data layer is that it says we're going to do, you know, one data model, but then like the fine text in the bottom. Basically it says that like, well, it's not going to be anybody else's data model because that's impossible. We don't have one.
And so, so like the subtext is that you're going to be then forcing the data model onto all the applications that are going to be writing on top of it.
Andrew Rodgers: [00:15:19] Right. Th I mean, so, so when you look at the private commercial solutions in this space, they do tend to be very opinionated about the data model. I think that's one of the things that is attractive about Vultron and, and how we've built our platform, you know, our platform and, and there'll be a little tricky today talking, separating some of the value we've added on top of Vultron in our platform versus what's built into Vultron itself.
I apologize. I will try to be clear about that. Yeah, Vultron in itself is not opinionated at all. It treats everything sort of natively as it finds it in the system that's being integrated. Um, and you know, for backnet, for instance, that just means you get back net device IDs, and, and point addresses and, and registers.
Right. Um,
James Dice: [00:16:07] Can you back up and, and let's, let's do dive into Vultron. Now we talked a little bit about it on the last episode, but assuming that people didn't hear that, what is
Andrew Rodgers: [00:16:16] sorry. Yeah. Yeah. Um, so, so Vultron is a open source technology project. It was funded by DOE led by Pacific Northwest national laboratory.
Really around the idea that DOE was doing all these really interesting. What they now call grid interactive efficient building projects, Gabs. Um, at the time I don't think it was that term was really a term of art. Um, I
James Dice: [00:16:42] think it was grid integrated buildings back then
Andrew Rodgers: [00:16:45] district. I like better sources, et cetera.
Um, and. They had this problem, which I've done a decent amount of work in research, uh, research computation, not, not in the national labs like James has, but in the more in the academic side, um, there's a lot of just wasted effort in like recreating. Simple boiler plate stuff that you need. Like, you know, for instance, if you're doing research on the absolutely best, optimization strategy, like every grant would have funding in it where you do the integration again.
So like, you know, everybody's like connecting the test, you know, uh, air handler to mat lab. In every single proposal, like it's the same work being done by different grad students. And like, you don't get, you don't even get the same grad student who's already done it once before. It's a new grad student who has to learn the whole thing again.
So DOE was really trying to say, like, let's build, let's build some infrastructure that enables this research to work faster and cheaper. So that's really where Vultron came from was, enabling. Rapid testing, safe, uh, operations. So like when you use Vultron as a two way platform, you know, it has all the semantics built into it for managing what priority things get written to in the backnet control or those sorts of things.
It has automatic rollback. So your application, if your application falls over Vultron will detect that and roll back to whatever the previous set point was before your application overrode it. Um, so it's got a lot of really cool features around that in the building space. And then it was more broadly, you know, not just the building systems, but designed to sort of bridge that gap between.
Uh, building automation systems back net mod bust, that sort of stuff that was in the building envelope and the grid scale, you know, utility distribution system protocols, like the MP3 and open ADR and, and those sorts of things. So it was really. Bill as an infrastructure platform that allows people to build applications rapidly and cost-effectively, um, by kind of solving some of these boilerplate problems that you just solve over and over otherwise.
Yeah.
James Dice: [00:19:03] And so how do you, you, you basically took it, got turned into an open source project through the eclipse foundation. I think that's been yep. And then people like you, there's probably other firms like you guys then build, build onto it and create your own product. Is that how it works?
Andrew Rodgers: [00:19:18] That's right.
Yeah. And so like for us, You know, our cloud platform is very much our product. There's not a lot. I mean, Vultron is involved in the data aggregation and collection, but then once it hits our platform, like the data Lake and the visualization and all that stuff is completely different technologies that we've assembled and put together into that platform.
Um, but we, we try very hard to keep the edge, uh, resources, pretty much vanilla Vultron. just so that. When one of our customers wants to deploy some sort of Vultron application, that's not ours, they have that capability. and that, that choice, again, back to, you know, kind of built into our ethos is giving our customers as much choice as we possibly can while maintaining, I mean, the, the risk here or the, the cost model, how you think about this as a business problem really is, you know, the more choice, the more risk.
And you have to pay for that risk somehow. So it's always, for us, it's always walking this really, you know, trying to find the best balance point for our we're giving our customers as much choice as possible while still like keeping it manageable, where we can offer a good product for a good price.
James Dice: [00:20:29] Totally.
Hey guys, just another quick note from our sponsor nexus labs. And then we'll get back to the show. This episode is brought to you by nexus foundations, our introductory course on the smart buildings industry. If you're new to the industry, this course is for you. If you're an industry vet, but want to understand how technology is changing things.
This course is also for you. The alumni are raving about the content, which they say pulls it all together, and they also love getting to meet the other students on the weekly zoom calls and in the private chat room, you can find out more about the course@courses.nexus lab. Start online. All right, back to the interview.
Well, the reason I wanted to dive into this to stay with you is I feel like this. Advanced supervisory control piece is hot, right? And the independent data layer is hot and they'll try and you know, is a heat tool for both of those things. Right. And. I wanted to ask you around, like, why would someone build their own piece, like proprietary black box inside the building to do that two way integration and why shouldn't they, I guess, and you just kind of, you said it, but why don't you say it like to the, to everyone else?
Uh, why shouldn't we
Andrew Rodgers: [00:21:38] do that? Yeah, I mean,
I would say there are good reasons to do that. And there are good reasons to not do that. I wouldn't say that, like, you know, if your business model is a traditional business model in this space of capturing a customer and like, Tying them down as hard as you can and giving them as little choice as possible, then like, yeah, write your own.
Right. Like use the most obscure stuff you can because that's how you're going to like, keep, keep them locked. Um, but for her calls it, lock them and lose them. Yeah. Yeah. Yeah. He does. Uh, so for the rest of us though, like who are trying to maybe see a better future in this space, you know, I would say this.
And I, I do say this to the Vultron team at PNNL, um, I'm, you know, one of the most active community members outside of the lab system, for sure. I'm also one of the ones that's like first to question every, but you know, there are things in the Vultron project that I don't agree with. Like, I wouldn't have done them that way.
I don't like them. I don't, you know, I don't think it's the right way to do something, but. The value of being in an ecosystem where there are other people using that technology, um, where we have sort of this plausible, presentation to our customers that like, yes, we're a small company, but we're using this technology that is open and, you know, there are other providers for, so if you decide that, like we're not.
Delivering you can fire us and you're not left holding a bag of some proprietary crap that you can't do anything with. And that value, you know, again, it's a very different perspective, right? Like, of course you want to tell your customers that if they fire you, they're going to be left. You know, they're going to have huge amounts of costs to recreate your so, you know, that's why I say like, From my perspective.
And I think maybe from the reality you would like to see in the world, no, no one should be doing it that way, but there, if that's your business model, that's your business model. And I mean, I'm not here to, I'm probably not going to convince you otherwise. And, and since that super as this model.
Yeah. So, um, but I would encourage you to investigate maybe like maybe we should be working on delivering value in a little more, um, Again, interoperable way. Like our goal is we solve the data aggregation, putting the interfaces together to allow control this advanced, you know, the advanced supervisory control.
We solve that problem really well. We don't compete with. Fault detection companies like we would, we love it. When our customers take our platform, bring the data back to a common layer and then hire a fault detection company to make sense of that data or deliver some, you know, direct value. And then turn around and say, Oh, well, our AFTD company, like they do everything really well, except for this one weird system that like, we're the only ones that have, because, you know, we made dumb decisions in 1980 or whatever.
Right. And then they, but then because they have this independent data layer, they have the choice to say, Oh, well, we're going to invest, you know, a few months of development time with our team to build something special that does that for us. And we have that flexibility. Whereas if they had just went with.
You know, an AFTD vendors, data collection tool that kept everything in a closed ecosystem and wasn't open and accessible. They would not be able to unlock that value. So that's where I think the independent data layer really starts to, you know, start to shine. And who
James Dice: [00:25:12] are the other people that are in the community, the Vultron community, besides the PNNL folks who, who is the people contributing
Andrew Rodgers: [00:25:19] back to them?
So there's a core user group. It pretty much each national lab that, that obviously, um, that is still using Vultron and working with Vultron, um, companies offering Vultron as a service, There's not that many of them, I would struggle to, to name one, who's like doing it in a way that's not, um, competing with any of the existing players.
Like there are companies that do fault detection and then they use Vultron under the hood to do that. There are companies who are doing, you know, completely managed building optimization services using Vultron. Um, but. From, we are probably some of the only ones who are out there, like selling it purely as the platform play.
Um, yeah.
James Dice: [00:26:08] And how do you guys sell it when it's just, it's not your product added on top. It's an open source piece of software that you're setting up. How does that
Andrew Rodgers: [00:26:15] work? The way I like to look at it is, you know, Anybody can go download Vultron off, get hub, start running it on their computer, run it, you know, install it on a raspberry PI if that's what they want to do.
Um, anybody can do that. What we bring is tools that allow us to deploy faster, collect data, better, more efficiently. Uh, monitor for changes in the network, monitor the hardware platforms we deploy, um, and maintain the health of those hardware platforms. We're really selling a service around reliable data acquisition.
That Vultron is a plays a central role in, but it's really the suite of, of, you know, the monitoring main, maintenance, all that stuff that we're offering on top of Vultron is really where our value comes from. I see.
James Dice: [00:27:10] Let's talk about the grid interaction piece of it a little bit. So Vultron has that, I mean, that was one of the intentions when it was created, it was just to be used for a building to interact with the utility.
Right. Um, what's the status of that today? Uh, and what are you seeing on, on real projects? Because I think where I'm at is. You know, leaving Enrail six months, seven, eight months ago, however long that was, I'm kind of out of that world, but like there's still this like opinion of buying where it's like, we're talking about a lot of hype for this and well-reasoned hype.
Like we need to decarbonize the grid in order to do that. We need to have load flexibility totally. 100% behind that. But what we haven't talked about is like, What buildings are out there doing it and why are they doing it? And does a building owner care about interacting with the grid? And so I guess what I'm asking you is like, give me an update on like load flexibility in grid interaction and 20, 21, like where are we at?
Andrew Rodgers: [00:28:12] Yeah. So, I mean, Vultron has been being used for, for. Pilot demonstration projects with utilities in this space since its inception. I mean, I, you know, probably day one with Vultron was that was the actual use case for it. Um, what we see is utilities starting to talk to us and say, Hey, we've been working with, you know, Oak Ridge national lab, Pacific Northwest national lab for years now on these pilots.
Can we get this thing to scale? Like, is this ready now? Can we actually do this? Can we start seeing aggregate benefit that is going to justify the investment? Um, at the same time, I think you're seeing things happen in buildings that are, you know, in the, the actual owner market and buildings that are pushing, you know, local law, 97 in New York, right?
Like, It doesn't directly affect grid interactivity, but because it's making people go out and touch these buildings and in P you know, do data collection, do, a lot deeper analysis of how energy is being used, then you start to see the movement that allows, you know, demand response or, you know, active load management.
To be kind of shoehorned into existing projects. Like, no, I think, I think we're, I think there's a, there's a huge shift, but I still think in 2021, no one is going to go out and say, Oh, it is worth it for us to go deploy all new infrastructure in our portfolio to achieve a load flexibility. Like nobody's going to do that.
But when it's, Hey, We're going to install vehicle chargers because our tenants are demanding electric vehicle infrastructure to install electric vehicle infrastructure. We've got to put 200 KW KWH a batteries in because that's the only way we can do it without completely upgrading our utility service, which is going to cost a million dollars.
That's when you start seeing, Oh, well for, you know, an extra tens of thousands of dollars, we can use this for demand flexibility. We can do our peak shaving. Like we'd get all these other ancillary benefits because there's already capital moving. I think that's where we'll see the most adoption on the building side.
The utilities are actually really interested in this stuff. On the residential side as well. So we're working with a few utilities on residential demonstration projects where you start asking the question, okay. People are putting, you know, Tesla power walls out there and rooftop, solar and electric vehicle in the garage and a water heater that's connected and you know, a smart thermostat that's, that's enabling control of the HVC.
Like if we start aggregating a hundred houses, a thousand houses, 10,000 houses, like. We can actually start making a dent in peak demand. Um, so I think, I think we're approaching market readiness. I wouldn't say we're like at full market readiness yet, but I think the market is in a much different place than it was two or three years ago.
Um,
James Dice: [00:31:22] and so w what does the stack look like from a Voltron perspective with that aggregation? You got a local node in each of the houses or in each of the buildings, and then you have
Andrew Rodgers: [00:31:31] one aggregator side. We've done. We've done some demonstration projects where, um, you know, Vultron is just operating in the cloud, kind of being an integration layer between a bunch of different cloud API APIs.
But when you start talking, especially in rural, America where things like battery storage can make a huge difference in electricity, reliability on long distribution lines, right? Like you put a few. Tens of KWS of our KWH is a batteries out there. And, you know, a storm falls a tree across the line somewhere and you can keep people's like critical services up like that can have huge impact.
I mean, Texas, Texas, this winter, I think is a great example of, Such a big failure at a massive scale that like the communications infrastructure failed too. So like, even if you had smart thermostats, it didn't matter because if they're going through a cloud round trip, they weren't getting out.
Um, right. So, so there's this desire on the utility side to really explore more resilient applications, right? So like if you've got a battery attached to a house and you've got solar panels on the roof, Can you Island that house and, you know, shut off all the non-critical loads and keep the refrigerator and the temperature in the house above freezing for, you know, 48 hours.
Like, and for that you need edge compute capacity. You need edge integrations to, to really do deliver that in a way that's fault tolerant. Totally.
James Dice: [00:33:05] Very cool. So let's talk a little bit about, Vultron sort of in context to the other ways of doing things. So we, we, we talked about the fully open model.
What are the other ways that like the independent data layer is being provided today? Like one of them is that I have, I'm curious about, you've talked to Mike Berman about SmartCore. Um, I'm sure. Like how, how does ultra and compare
it
Andrew Rodgers: [00:33:28] to smart core? So Mike and I had a really great conversation.
I think both of us are excited to work together. It's just finding the right application. Um, but smart core is much more focused on. Like day to day operations of the building and interaction of, um, sort of, yeah. User interfaces in Vultron has almost zero focus on that. Right. Vultron is about backend services, enabling, you know, these complicated control algorithms that need data from a lot of different sources and need to right.
Set points back out to a lot of different systems. Um, And I think the other thing smart core does not focus on at all is that grid integration piece. So Mike and I had talked about some applications where smart core might be doing all of the systems, uh, dashboarding and, and, you know, HMI, all that stuff in a building.
And then there would be a Vultron instance that it would communicate with as well in order to manage the grid, to building an interaction. Um, so I think, I think. Specifically with SmartCore. I don't think Vultron has, uh, Vultron is just a very different approach. Um, and maybe you can get to the same place, both ways, but I think depending on who you are, where you're coming from, you might would pick one of those tools versus the other versus necessarily like the specific problem you're trying to solve.
Other than like the UI stuff. That's all smart core. And, and H and Vultron just really, doesn't doesn't even try to address that. Yeah.
James Dice: [00:35:02] Got it. What about the other alternatives, I guess, doing sort of the open independent data layer, uh, on Voltron?
Andrew Rodgers: [00:35:10] Well,
I don't know that there are a lot of, there's not a lot of alternatives for a truly open, independent data layer right now. Like we see a lot of vendors doing their own, Models their own sort of proprietary solutions. And the question I'm always left with kind of, or, or the question I'm always asking when I review those systems is, are like, Where is the choice?
Like what happens if I want to separate myself from this company in the future? Where, where does that leave me? Um, you know, do I have a path where I can, you know, take my data and reimplement, another set of infrastructure that allow me to collect the data I need? Like, does that mean I need to completely redo all of my, my applications.
so that's. Where I think Vultron has sort of a unique focus. Um, there doesn't seem to be a lot of market pressure otherwise for these companies to really be open. Um, the other thing that you know,
it seems like there's a lot of effort on the companies who are saying, like, we're delivering an API for your building, those kinds of value propositions. Yeah. Yeah, they're controlling all of that and you know, if, if you're just using the model they have, you don't have control or choice in, you know, if you want to add additional features to that model that are specific to your use case, It's not clear to me how someone can, can deploy something that implements a, you know, somewhat proprietary model and decide you wanted to swap that with something else.
It's not clear how you would go about recreating that. Um, you know, data collection infrastructure in a way that, that, you know, was self-actualized
James Dice: [00:36:56] absolutely. Especially when you start to talk about connecting a bunch of different applications to it. So if you just have, you know, if you've pictured just a point solution today, you have one application connected to the data layer.
Right? That's great. It could be pretty easy to swap out components there, but when you start to talk about like the true value proposition of the independent data layer, when you have, you know, all these different applications plugged in, right. Uh, and now you're swapping out the infrastructure piece that, yeah, that does sound very, very difficult.
Um, interesting. So how about on the, let's go back to Vultron a little bit the advanced supervisory control today, like. I think I'm always wanting updates from the field because, um, it's still an area where I think, I think you have about 50, 50 of like there are a ton of startups that obviously believe that it's the future.
Right. I think on my vendor landscape, there's, it's up to like 15 to 20 now, um, of people doing this, Cloud application providing advanced supervisory control. And then there's the other 50% of the people that it's like never going to happen. Right? Like this is absurd, right. That we would provide control from somewhere else besides the automation system or the system that was designed to control the systems.
So. Well, like, can you provide an update on, on the, kind of where the market's at today? Um, I think this is an important question because in order to make things like grid flexibility happen, right? Like we talk about, Oh, we need grid flexibility, but in order to make it happen, we as an industry have to kind of get on board with the advanced supervisory control.
It's just the fact.
Andrew Rodgers: [00:38:40] Yeah. And, and I mean, I. I guess I just have commitment issues in all of these questions. Like you'll ask, well, is it this way or this way? And I'm always like, well, we just decided and just support both because it turns out like, you know, that's, that's where we, so, so when we talk about data modeling, like we are not opinionated about data modeling, we built out infrastructure to support whatever models our customers bring to us.
We talk about. You know, edge control versus cloud control. We have customers who were saying, Hey, we are actually already have an independent data layer. You know, we did it early. We were very advanced. Uh, but we didn't get to way control and like our vendor's not going to do that.
Or the system we built out internally is not going to do that. Can you go deploy Vultron and just give us an API for riding back, like these key set points from your cloud. Like that's, that's like super easy for us to do, but then if you want to like drink the Vultron Kool-Aid maybe you say, Hey, can you put Vultron in our buildings?
And give us a platform for us to deploy these applications that we want to develop that are doing this automatically and run at the edge. And the answer there is yes, as well. Like that's, that's why we kind of focus on doing this, um, you know, sticking to as close to vanilla Vultron as we can.
And those edge applications is so that we can take these packaged, you know, what vault in Vultron world as an agent, um, which isn't, you know, basically an independent application that runs in the platform context with access to the data that you provide it. Yeah. Via sort of an access control list. Um, so if you're familiar with deploying things into cloud environments, You know, you can choose very granularly.
What data, a particular application has access to. And there are open source. They're really great open source things that the labs have built that are available as Vultron agents. There's, um, there's one called intelligent load control that PNNL has built out. That's all around sort of doing peak demand, limiting with comfort in mind.
So you can kind of prioritize spaces, um, in your building and then set a target, load profile that you want your building to maintain. And it will, kind of. Shed load in a weighted way that tries to keep your priorities zones happy. Um, and then maybe we'll give up on shedding load in order to maintain a certain like, threshold of, of comfort that you're not willing to go outside of.
Um, and so that's like a Vultron agent that you can, as again, open source, you can download it, you can install it, run it on your building and achieve, you know, peak load reduction. Pretty pretty straightforward. so that's like the fully drinking, the Vultron Kool-Aid way to do it. It's running in the edge.
It's resilient. It's kind of giving you all the value proposition that people would say that your existing BAS gives you. But it's still a supervisory system that can pull in data from a lot of different sources instead of just the BAS itself. And you get cloud or, you know, external connectivity oversight.
So you can easily connect, bring all that data back to one place while the actual decision-making is happening at the edge. Um, one of the things, and I guess this is my experience in the industrial world, but I was amazed when I started. In building automation and found out that a lot of these startups you're talking about who do say the independent data layer, like a lot of it it's like they install a VPN and then they have like a backnet client in the cloud, somewhere talking to the BAS over this VPN link.
And if the connectivity goes down, you don't get data. Like there's all kinds of reasons not to do that. and it just in an industrial world that would have never flown because You don't want to give up having your data just because your internet connections down. Um, you proxy that somewhere locally Vultron does all that for you.
So not only can your edge device be sitting there running some optimization strategy, it's also collecting the data, buffering it locally, forwarding it to the cloud. So you still have sort of a portfolio wide supervisory view of what's going on, but you can actually deploy those control. Algorithms directly to the edge device.
Absolutely.
James Dice: [00:43:07] Interesting. That that's a good enough for me like that you had me at like our clients are asking for this, like, like I feel like we just need more case studies around actual building owners investing in these types of controls, because I still think. I again, I already said it. I still think that we're, we're not quite as an industry decided that this is a good idea at this point.
Yeah. Yeah. And I've heard several people that like several building owners, large building owners that have, I just heard about recently that are investing in this approach and we just don't have enough public case studies on it yet.
Andrew Rodgers: [00:43:45] Yeah. I mean, it's, One of those challenges, because if you're really good at it, it's definitely a competitive advantage.
Um, and you know, if you're a big either portfolio, owner or manager, you don't necessarily want to talk and give away all of this, you know, great ways that you're improving your return on your investment. Um, so it's, it's a little tricky. I, you know, I still think, you know, Joe Gaspar, Tony always makes this point that like, CRE is the hardest place to do this.
And you know, Â REITs are really hard to do this in because everything is driven by this value proposition and return on investment. Um, but I do think from what I've seen is like some of the most interesting things are being done by the private, um, private and commercial owners instead of the, the campuses and universities and such.
I think, I think, I think it's that three 3,300 thing. Um, that drives that, you know, Â and then like the future's already here. It's just not evenly distributed. Like somewhere between this to stay between those two cliches is why they're a few property owners doing really cool stuff today. And we don't hear about it a lot. Right, right.
James Dice: [00:45:05] Totally. Well, this has been fun.
I think that's, that's the end of my current round of questioning for you. Did you have anything else you wanted to kind of get off your chest? This,
Andrew Rodgers: [00:45:15] this round? Uh, I don't know. Um, this was, this is always good to catch up. I appreciate you having me for sure. Um, you know, I'm always happy to talk to anyone about Vultron.
it's, it's very exciting to me as a technology. Um, I've just, I I've talked about, I think you'll, you'll see, like I post things on LinkedIn a lot and I talk about market transformation forces. I think Vultron has that potential, not necessarily that everyone's going to adopt Vultron, but that, that, open platform that exists that does some of these functionalities is just, uh, a pressure point.
Like it's our pressure source, I should say, to move the needle and. you know, that may mean that like know, Mike Ruman at SmartCore has, you know, CS. Voltron's like, Hey, if they can do it, I can do it. Um, it might mean that, you know, someone has to have to bid against a Vultron solution and it encourages them to change something about their product that makes it more open.
Like there's all kinds of ways. you know, The existence of something like that can change a market. Um, I think I posted recently something, I was really amazed, like this really great industrial controller, based on the Arduino platform, very low cost, really cool features, and to see how the accessibility of embedded development of industrial control using open technologies.
All of that's been moved by like some crazy professors in Italy who wanted to make embedded development easier for their staff. It's like Arduino as a market force is a really interesting story. I think it has a lot of parallels in what something like Vultron and smart SmartCore can do in this sector.
James Dice: [00:47:06] Amazing well to keep, keep getting updated. Uh, I always love your LinkedIn posts and please keep it up and we'll have to talk more about, uh, uh, Andrew on interoperability. That's what I'm going to call it. Uh, we'll have to talk about it after we, after we hit stuff here. So thanks for coming on the show.
Thank
Andrew Rodgers: [00:47:24] you.
James Dice: [00:47:25]
All right, friends. Thanks for listening to this episode of the nexus podcast for more episodes like this, and to get the weekly nexus newsletter, which by the way, readers have said is the best way to stay up to date on the future of the smart building industry. Please subscribe@nexuslabs.online. You can find the show notes for this conversation there as well. Have a great day.
Head over to Nexus Connect and see whatâs new in the community. Donât forget to check out the latest member-only events.
Go to Nexus ConnectJoin Nexus Pro and get full access including invite-only member gatherings, access to the community chatroom Nexus Connect, networking opportunities, and deep dive essays.
Sign Up
Nexus Labs exists to move the smart buildings industry forward, without the fluff.
On Wednesday, we send out the renowned (and free) Nexus Newsletter.