Teaming Up To Enable Digital Transformation & Unified Namespace

Speakers

Kevin McClusky

Chief Technology Architect

Inductive Automation

Arlen Nipper

President & CTO

Cirrus Link Solutions

Benson Hougland

VP

Opto 22

Mark French

Director of Design Consultation

Sepasoft

James Burnand

CEO

4IR Solutions

Are you wondering how to best proceed in your Digital Transformation journey? Join us for an exclusive webinar featuring Inductive Automation and key members of its partner ecosystem – Opto 22, Sepasoft, Cirrus Link, and 4IR Solutions – as they showcase how to accelerate and simplify Digital Transformation through a collaborative, integrated approach.

This webinar will highlight the benefits of using all partners within the Inductive Automation ecosystem to achieve seamless digitalization and a Unified Namespace (UNS) with reduced complexity and time.

As a special feature, the webinar will include a live demonstration of a real-world digitalization application, where you’ll see how these systems work together to address common challenges. You’ll experience firsthand how a combined solution from these partners empowers users to implement and leverage UNS for streamlined data access, real-time insights, and improved operational efficiency.

• Learn how integrating solutions from multiple ecosystem partners speeds up your Digital Transformation.
• See how UNS enables centralized, accessible data across the operation for quicker, data-informed decision-making.
• Find out how this ecosystem supports scalability and adaptability, making it easier to respond to changing needs.
• Discover how IA, Opto 22, Sepasoft, Cirrus Link & 4IR Solutions combine their strengths for a best-in-class, connected approach to digitalization.

This webinar is perfect for industry leaders, automation engineers, and Digital Transformation strategists aiming to maximize their digital investments and drive efficiency. Don’t miss the chance to see these tools in action and learn how UNS can transform your operations.

Transcript:

00:00
Kevin McClusky: Hello everyone. Welcome to our webinar "Teaming Up To Enable Digital Transformation and Unified Namespace." We're so glad that you can join us today. I'm Kevin McClusky. I'm the Chief Technology Architect here at Inductive Automation. And joining me are Benson Hougland, the Vice President of Marketing at Opto 22; Arlen Nipper, President and CTO of Cirrus Link; Mark French, the Director of Design Consultation for Sepasoft; and James Burnand, the CEO of 4IR Solutions. The panelists thanks for being here. Can you each take a minute to tell a little bit more about yourselves and your company? Benson, let's go ahead and start with you first.

00:41
Benson Hougland: Sure. My name is Benson Hougland with Opto 22. I'm responsible for product strategy and marketing here at Opto. I've been here for about 30 years and in the industry for about 40. Opto 22 kind of heads up the edge portion of our webinar today. We are manufacturer of industrial automation hardware and software, all made right here in Temecula, California, made in the USA. And appreciate your opportunity to be on this webinar with you guys.

01:12
Arlen Nipper: Hello, everybody. Arlen Nipper. I am the president and CTO, Cirrus Link Solutions. I've been doing industrial automation for about 47 years now, and 25 years ago I was very fortunate to work with Andy Stanford-Clark and be one of the co-inventors of MQTT. So Cirrus Link is the only strategic partner with Inductive Automation, and we work five days a week, eight hours a day, in creating MQTT technology modules that plug into the Ignition platform.

01:47
Mark French: Hi everyone. Mark French here from Sepasoft. I'm responsible for training and sales engineering here at Sepasoft. Sepasoft is a software vendor that provides MES capability on the Ignition platform, as well as, well, recently, our upcoming releasing an analytics platform that we'll be showcasing today. James.

02:11
James Burnand: Thanks, Mark. So I'm James Burnand. I'm the CEO of 4IR Solutions. 4IR Solutions is an Inductive Automation solution partner with a focus on managed infrastructure. We've been in business since 2021, and what we do is we try to make it easy for folks to deploy and operate their OT infrastructure. We do that using our platform-as-a-service FactoryStack. We look forward to describing that to the team today. Thank you.

02:37
Kevin McClusky: Thanks, James, Mark Arlen, Benson. These folks are some of the finest folks in the industry, if I say so myself. We have a great relationship with all of our partners, and it couldn't be happier to have these powerhouses of the industry here with us today. So we'll turn off the cameras to save a little bit of bandwidth, but we wanted to let you all see our faces as we got started here too. We're real people, and we care very much about this community. So without further ado, if you're not familiar with Inductive Automation just in general, here's a few notes at a glance. We've been in the industry for 22 years. We have our 350 employees that are sitting inside the US and Australia. Were used by about 65% of the Fortune 100, 44 or so of Fortune 500, and a really diversified customer base across many industries.

03:30
Kevin McClusky: Thousands of Ignition installations, tens of thousands of Ignition installations in over 100 countries. We also invest heavily in our integrator network. We have over 4,000 integrators worldwide who are part of our integrator program. And if you're wondering what Ignition is, we like to say it's an unlimited platform for SCADA and so much more. In fact, Ignition enables you to connect, design, deploy basically any industrial application you need. So that'll include SCADA, HMI, MES, IIoT, and the list goes on and on there. It acts as a central hub for everything on the plant floor and beyond. Really, a cornerstone of Digital Transformation technologies on the industrial side is something that we see a lot of companies doing these days, but it can be used in really small applications too. So it scales from small to large. It's web-based, web-managed, web-deployed, an unlimited licensing model.

04:26
Kevin McClusky: It's cross-platform. And of course, as you have to inside this industry, it offers industrial strength, security, and stability. We pour a lot of effort into those pieces. We also work with several technology partners who help solve many digital challenges to get more power from Ignition. So our strategic third-party module partner, Cirrus Link, creates IoT modules that let you set up a secure MQTT message-oriented middleware infrastructure that increases data availability. You just heard from Arlen, who leads up a lot of efforts at Cirrus Link. Sepasoft, 4IR, and Opto 22 are all in our Solutions Partners Program.

05:11
Kevin McClusky: As you've noticed, these are the folks who are presenting here today. 4IR provides an easy way to deploy Ignition and its partner ecosystem in the cloud via fully managed solution. They manage updates, monitor system health, provide 24/7 support for your infrastructure running inside the cloud. Opto 22 manufactures hardware and software for industrial automation, remote operations monitoring data acquisition. They specialize in electronic equipment, including Ethernet IO, edge programmable industrial controllers. Today we will be demonstrating how Inductive Automation and these four partners combine their strengths to enable Digital Transformation. And we'll show you a pretty neat use case here. So, speaking of which, I wanted to bring Benson into the conversation. We'll all be bouncing back and forth throughout this webinar. Benson, can you talk a little bit about the use case we're going to be looking at here today?

06:07
Benson Hougland: Yeah, sure thing, Kevin. So in this use case, our customer is the owner of a chain of convenience stores called Epic C Stores. Now, it's worth noting that while we're featuring convenience stores as our use case for this webinar, recognize that this is just a surrogate use case of many industrial automation applications. For example, in a C store, you connect to physical assets like pumps, levels, flow, machine status, and more. You collect that information, process and control it, visualize it, analyze it, and much more, just as you would in a more traditional automation scenario. Now, what's important is that what services the C store provides and how we can increase operational efficiency at scale. Speaking of scale, our customer wants to do just that. Their goal is to expand from just a few stores to 50 very quickly and ultimately become the largest C store chain in just five years.

07:02
Benson Hougland: Now, as a side note, the US convenience store industry is estimated to be an $800 billion market. So as Epic C Stores embarks on its Digital Transformation initiative, they must identify specific objectives to provide the organization's stakeholders with the tools and the data they need. The list is long, but generally they need real-time visibility of their entire operations by connecting, collecting, modeling, and analyzing their operational data. Then they wanna optimize their inventory to meet demand while keeping costs low. And they also wanna provide local control and visibility. Further, they'll need to model and publish their data where the source of truth exists at the edge in the store. This will allow other applications to consume that data and perform tasks like long-term history, machine learning, analytics, and much more. All of this must be done over a strong, cyber-secure architecture to eliminate unauthorized access, but while still providing secure, authenticated remote access for troubleshooting and maintenance.

08:08
Benson Hougland: And ideally, open technology should be used throughout to avoid vendor lock-in. However, one of their most important considerations is using the right tools that will quickly scale to meet their growth objectives. So the company is growing fast and opening new stores frequently. So it needs a technology solution that helps it meet its goals and objectives. Our solution for this is to build a multi-site application with the following capabilities. They'll be able to monitor all of their stores in their fleet from a central location and within each store, monitor food storage units, fuel levels, car wash supplies, local weather, lighting, store transactions, and more, all in real time. And also very important, they'll be able to commission new stores on the fly.

08:54
Kevin McClusky: Like so many companies today, Epic Stores that you just saw need to do more with their data in order to achieve their goals. And of course, this is not a real company doing real things. This is the demonstration, but you can see how this would fit into your businesses. Most companies today do need to do more data, but more with their data in order to achieve their goals. And scalability is very important. In these Epic stores, they need to increase automation to make data more accessible so they can be more efficient and streamline their operations. They need to use data to improve customer experiences so that their customers will stay loyal. They need to leverage edge and cloud technologies, data analytics, and emerging technologies like AI, machine learning, and predictive maintenance. They need data to make better business decisions and to be more adaptable to market changes like gas and food pricing changing on the fly.

09:50
Kevin McClusky: They need to make cultural changes inside their organization that embraces a wider availability of data within the company and an understanding of what you can do with that data and everything just described there. That whole effort can be summed up in two words: Digital Transformation. There are various ways to implement Digital Transformation, but one of the best ways is to set up an accessible single source of truth for data known as a Unified Namespace. Digital transformation is normally multiple things. An important part of those initiatives is supported by this Unified Namespace. The Unified Namespace, or a UNS, acts as the foundation of your Digital Transformation initiatives. In a UNS, data from all areas of your company are organized into a fully accessible, secure architecture that captures processes and responds to events from decoupled devices and systems. As mentioned, we're going to show you a demo of this. We'll talk through some of the tag structure and some of the other pieces of this in just a few minutes here. The first step in building out a system like this is the infrastructure. So I'm going to turn this over to an expert on infrastructures in general. So James, over to you.

11:07
James Burnand: Thanks, Kevin. So 4IR's mission is to make it easy to deploy, operate, and manage your OT systems using our platform as a service, FactoryStack. So for today's demonstration, we've set up an infrastructure in the cloud that's being used by all the partners today. So this includes Ignition, the Cirrus Link distributor, MQTT broker, the Cirrus Link MQTT client Engine and Transmission, a dedicated database, a private GIT source code repository, and all of the associated cloud services and configuration to make all this work smoothly. We take care of creating and renewing SSL certificates, providing DNS, performing backups, security monitoring, enforcing multifactor authentication, and upgrading not only Ignition, but all the other applications and components that are a part of this architecture. This system, like all FactoryStack Systems, is connected to our central management plane and is monitored and managed 24 by 7.

12:05
James Burnand: So we thought it was worth highlighting how that's different than when you deploy a virtual machine or another method to deploy Ignition in the cloud. So what we wanted to do was describe that to you in a little bit more detail. At 4IR we use a technology called infrastructure-as-code. This allows for us to programmatically deploy and manage instances of FactoryStack wherever they're deployed. This starts on the left side of this diagram that's showing Docker repositories. These are fully supported images provided by Inductive Automation, Microsoft, Postgres, Grafana, and others that provide a Docker-based container image to be used for deployment. 4IR then takes those images and hardens them to meet the requirements of our customers environments. We then digitally sign those containers and make them available for use in our environments. We also, on the top of this diagram, you'll see we get updates from vendors and cloud providers that are not in a Docker format that also have to be included inside of our systems.

13:06
James Burnand: So all of that is pulled into our management plane, and then these changes are included as part of our release process. Our release process then allows for those changes to be pushed out to customer environments when we trigger them. We do not automatically push updates into customers' environments. Rather, we work with each client to define update Windows to ensure that their software stays up to date. As you can see, that's not where the arrows in the boxes stop. We've also built in hooks and integration into our monitoring and alerting system. This ensures that every FactoryStack that is out there can consistently report its health back to us, and we can proactively respond 24 by 7 if something isn't working as intended. So this diagram covers off the cloud portion of how this demo is being handled, but I'd love to pass it over to Benson to talk a little bit more about how the Opto 22 groov devices handle the edge. Benson.

14:00
Benson Hougland: Right on, James, say Arnel, would you share our C stores live video with our audience? Okay, that's cool. So at the edge, our C store has many IO or inputs and outputs, to sense and control many aspects of the store, such as the fuel systems, refrigeration, lighting, HVAC, energy use, and the car wash. So we wired all these up to the groov EPIC edge programmable industrial controller through its IO cards and developed a real-time PLC program to control the store's various subsystems. Then we fired up Ignition Edge running on the EPIC to collect and model all of the store subsystems into UDTs, or user-defined types. From there, we've built Perspective visualization screens with UDT templates so as we scale to more stores, screen development and deployment are simple and repeatable. Then those same UDTs are published outbound and report by exception securely to the FactoryStack infrastructure where an MQTT broker runs.

15:02
Benson Hougland: Now, any applications with proper permissions can access the C stores real-time data directly from the broker in the cloud. Now this approach dramatically improves the cybersecurity posture of the whole system, and EPIC's built-in firewall blocks all inbound access to this store from the internet. With EPIC's built-in open VPN client, which also connects outbound to a VPN server within FactoryStack, we have remote, secure, and encrypted access to the store on demand. I should note that all of the edge of functionality I've just described occurs on just one single manageable and secure system, the groov EPIC. Now we're also using MQTT and Sparkplug in the solution, and it has tremendous advantages. To explain this further, I'll pass it over to Arlen with Cirrus Link Solutions.

15:54
Arlen Nipper: Thanks, Benson. One of the extremely unique aspects of the Ignition platform is this powerful tag management system that takes pretty much obscure registers from numerous machine protocol sources and turns them into contextual, humanly understandable tags and UDT models. The modules from Cirrus Link tap into all of the tags and the contextual information within Ignition and publishes them into an enterprise infrastructure using a secure open protocol called MQTT. Now on top of MQTT, the Sparkplug specification is the layer which defines how best to publish that data and has been influential technology in enabling UNS across our customer base. Because of the popularity of the UNS, we've added some new UNS tooling within the MQTT modules. One is the ability within MQTT Engine to visualize all of your tags within a complete UNS structure. The other is the ability to individually publish tags using our UNS tag publisher. And with that, I'll pass it back to you, Kevin.

0:17:14.4
Kevin McClusky: Thanks. So where is Ignition in this overall architecture? Of course we need to talk about that piece here as well. So Ignition is sitting right on top of the solution, and it gives you the visualization of that process. Ignition enables you to build and view practically any kind of industrial application inside web browsers, mobile touch panels, workstations, and more. And if you don't know much about Ignition, I'll just mention a couple of other things here. It lets you connect to all your devices with unlimited tags and connections at a very reasonable price, and you can build a highly scalable architecture inside Ignition. Everything from smaller centrally managed server architectures up to larger distributed architectures where one gateway manages multiple agents or in scale-out architectures where workload is distributed between multiple gateways. After you have a visualization and data coming into a central system, you might wanna take the next step in doing data analytics as well. You can do some of that inside Ignition, and you can do some of that inside external platforms. Mark, we'll talk a little bit more about the latter options, so I'll hand it over to you, Mark, for that piece.

18:25
Mark French: Thanks, Kevin. Yeah, I'd like to highlight some features of the SepaIQ analytics platform and explain how this product plugs into the ecosystem of partners presenting today with the demo app. So leveraging Sparkplug, SepaIQ is actually subscribed to C store topics in the Ignition-hosted MQTT broker running in FactoryStack. This allows SepaIQ to leverage the infrastructure that's already been described. These features allow us to store data, calculate, add additional context while streaming that data in, and then return analytics in stored on demand or subscription fashions. So this process of streaming data in or out... we're adding an additional contextualization layer. This provides a lot of value for end users, especially in manufacturing, as they spend a considerable amount of time on this contextualization activity before they can leverage the data with machine learning and AI tools. So in the present example, we're streaming data to a single server, but server clusters, leveraging stateless SepaIQ instances, can be deployed, allowing users to scale up applications very easily. In case some of that language might be new to you. When I say SepaIQ instances are stateless, I mean we can add or remove instances, and they're ready to do work immediately without importing a lot of historical data or execution information from past runs.

19:47
Mark French: Anyway, scaling for load without tedious configuration is a strength of the Ignition partner ecosystem and a benefit for our mutual customers that supported here by SepaIQ. Now, the calculations that I'll be showing later today are simple enough, but more complex calculations can be furnished. Customizing your own KPIs for your business or leveraging machine learning models and producing predictions. Let's take a quick look at a data flow on the next slide. Oftentimes when we're talking about data moving from here to there, we can get lost, and it can be a little abstract. So I hope this graphic helps. So as data comes in, typically through MQTT, raw values are stored off in a time series database according to the structure defined for the user's needs. In our current UNS focus, this could align and expand upon is ISA 95 or be completely distinct from existing manufacturing standards.

20:42
Mark French: So as the raw data is ingested, we can enhance that data with calculated values and other factors. This is indicated by the blue box in this graphic. These calculations could be simple things, little cleanup tasks like rounding or summing, or they could be complex like a multi-step KPI. This gives end users the ability to rapidly retrieve those calculations later in the context of the telemetry-style data that we're subscribed to for the C stores. In addition to telemetry-style data, we can also record transactional data. This could be through the same method like MQTT, or it could be something like a REST endpoint, Kafka message queue, database query, or a direct connection to Ignition.

21:25
Mark French: Again, when we're talking about UNS, the Namespace definition has considerable flexibility according to the needs of the customer organization. So we need to be able to accommodate that and relate data according to the desired labels. Now last thing I'd like to point out about this graphic is the ability to define your calculations on the way outta the system so requested analyses can publish a variety of data to suit your needs. Examples include, of course, the raw data and calculated data I already mentioned, could be predictions from machine learning, or we could even bring in and calculate off of master data from relational databases that are connected to SepaIQ. So that being said, I'd like to transition to the next slide. And just for the folks that might be new to the Ignition ecosystem, we're typically known for our MES modules that install directly on Ignition. So we don't just do analytics; we also do MES. So back to you, Kevin.

22:21
Kevin McClusky: Thanks, Mark. So this is the architecture of what the demo is going to be today. And I'll just walk through us one piece at a time. At the edge, like we talked about, the groov EPICs from Opto 22 are on the C stores, and they have the C store logic inside them. Each store is modeled as a single UDT, and those groov EPICs are connected up to equipment and streaming that information back; they end up streaming it to an MQTT server. That server is sitting in our case in the cloud as part of a larger infrastructure here. But that server can be anywhere. And that's part of the IIoT infrastructure in this case for this demo. The UNS as well is contributed to from everything connecting up to that MQTT server. We'll show that off in just a minute here.

23:14
Kevin McClusky: 4IR provides the infrastructure that's sitting inside the cloud. So you have Ignition in the cloud there. That could be Ignition Cloud Edition, or that could be standard versions of Ignition that are running in the cloud. Feature sets the same, or it can be between those. We'll run inside AWS and Azure, as you can see. And the important part here is this is where the Ignition screens that you're about to see are sitting. This is where the basic visualization is happening. This is what's serving out the clients and serving out any mobile devices that might be taking a look at this. It's a SQL database that's sitting inside that infrastructure there. And 4IR has a few additional pieces in their standard infrastructures that they spin up as well. To keep this simplistic, I wanted to show the main highlights and the pieces that are the most important for our demo here today.

24:07
Kevin McClusky: Now, on top of this is where Sepasoft and SepaIQ is happening. They're connected up to that same MQTT server and connected to the same UNS and pulling information from there. They're running Ignition locally and the SepaIQ application as well. So when Mark shows that off, you'll be pulling from the exact same data, exact same structure, exact same Namespace, that same UNS that's being used for the whole demo here. And just as a side note, other additional applications can connect up and use this UNS as well. So, for example, Cirrus Link has a bridge for Snowflake, and if you're connecting up Snowflake, that would connect straight to that MQTT server. There's 100 logos that I could have put here for showing off the other pieces that can connect straight up to that MQTT infrastructure and get access to this Unified Namespace.

25:04
Kevin McClusky: With that said, I'm going to hop right over to a quick demonstration, and I will see if... excellent. So this is Ignition running, and I'll switch back and forth between the architecture diagram and the actual demo here. But this is Ignition running where it says Ignition in the cloud there. This is running from that server. I happen to be a client that is connecting up there. And if I take a look at this right here, these are my tag providers, and this is my UNS, the Unified Namespace. Now, we had a lot of questions before this webinar, folks asking us information about how do I set up UNS, where does this actually live? We'll get to a number of those questions afterward. Also, as a side note, as you have questions as we go through this demonstration, please feel free to add those questions directly to the GoTo Webinar panel there.

26:01
Kevin McClusky: So you have a questions panel; please feel free to use that. Inside this UNS we have in our case Epic C stores that are connected here, but this UNS could be built out of anything. And so some folks end up building their UNS out of an ISA 95-style model where it might have a region, it might start with the country, or it might just start with the region, and then it might start with an actual location. So you could have US and then you could have California and some other things there. Other folks do it more organizationally from structure of the things that they're tracking. And this company, for example, might have multiple things in addition to C stores. There could be a lot of other information that's sitting here in the root of the UNS as well. And so the choice here was to have the C stores that are sitting directly in here. Under that section, it has a number of different folders.

26:54
Kevin McClusky: So there's, of course, supplies that are going to the C stores. There's transactions that are happening inside here that might be fed off to an accounting system. Stores themselves have a lot of real-time data that are streaming back and forth. And so this whole UNS we will go into in a little bit more detail here in a few minutes. But I wanted to give you a bit of background as to what we're publishing into and what Benson's pieces will be connecting to here. And quick view of some visualization of that. We have a couple of C stores that are sending near our offices here. This is one that's in Folsom. So we can see here that the convenience store has a lot of different features to it. This is the EPIC Panel that's happening right now that's on the C store and collecting data and sending it back. We have car wash, freezer, fuel systems. We can see how the HVAC is doing. And once again, we'll switch back to this in a minute. But this gives you a quick view of that data coming in actually being accessible, being able to be used, shown off, demonstrated, and then later on analyzed. So with all that said, Benson, I will go ahead and pass this over to you at this point. So I'm making you a presenter, and that should give you the ability to share your screen.

28:11
Benson Hougland: Great. Let's see. I'm gonna make sure I can get the right screen here, which is always a challenge. And let me know if you guys are seeing my screen.

28:20
Kevin McClusky: Yes, we are Benson. I think I would take just a moment to answer a quick question that came in. We've had a couple, but this one's a really quick one. "Just curious why you're showing the UNS inside Perspective instance. Is Ignition the UNS in this case?" And the answer is no, Ignition is not the UNS. In Perspective it's real easy to show it off. You can show it off inside the Ignition designer as well. We just figured that it would be a little easier to demonstrate in a way that was showing some of the tag values there. And I'll show a little bit more there. But the UNS is part of everything that's contributing to it and the MQTT server. You connect something else, you're going to get that exact same structure; that same UNS is going to be accessible everywhere regardless of if the client is Ignition or something else. Alright, Benson, back over to you.

29:09
Benson Hougland: Great. And then, let's see, we have one of the C stores up. And what you see there, of course, is the Model C store. And for one of the things that... For anybody say who went to ICC last September, these may look familiar to you. They are the same demo units that we put together for the Build-a-Thon. So these are real, real IO on there, real sensors, actuators, all of that. And Terry, if you'll go ahead and turn on an additional location, there we go. And this, of course, now is the C store here in Temecula. So one of the things I want to do is we'll come back over here to our map, and you can see that I've got those C stores already in there. But what I wanna do is add a new one. And because the customer's objective was to be able to scale quickly, I wanna show you how we're gonna be able to add a store real quickly down here into Temecula.

30:02
Benson Hougland: So to do that, the first thing I'm gonna do is actually log into that C store that's gonna be the darker one in your live video. And I'm gonna do that here with this screen and actually log into the EPIC. And I am doing this over a VPN connection, and it is secure. So I do need to put in a username and password because everything on the system is account-based. So here's my accounts, and I can go and manage my local users; one of those Benson, here I just logged in with. But those are accounts on the local device. You can also manage your accounts centrally using LDAP, and that might run also up in the FactoryStack server. So a lot of different ways to manage access to this device. As I described, we've got IO cards right there on the EPIC and it's tied to all the sensors, actuators, levels, temperatures, and so on.

30:54
Benson Hougland: And then we have a control program running in there as well that is managing and controlling all the subsystems in the store. Now, a couple of quick points: you'll notice that this is a completely web-based interface for managing the groov EPIC and all of its services. It also is where we manage all of our security. So I have my web server certificates and all the tools required to create certificates, upload, update certificates; all that's built in. And as I mentioned, we also have a built-in firewall. So for all the services that are running on the EPIC, I can then manage access to the EPIC through that firewall on various network interfaces. And speaking of network interfaces, very quickly we'll take a look at the status of the network on this system. And we have multiple IP network space.

31:45
Benson Hougland: This is within the store, so anything we're communicating is completely segmented from this other Ethernet interface. And this is my outbound access to the internet and therefore to the Ignition up on FactoryStack; the MQTT broker, the open VPN server, all of that is outbound. And then, of course, you see that my OpenVPN client has connected to the VPN server, and that's why from this screen I can get in here and manage this device remotely. Great. There's a lot more in groov EPIC that of course, I don't have time for in this webinar. So we're gonna jump right into Ignition. And indeed Ignition in this case, Ignition Edge is turned on and it's completely built into the system. So you literally just turn it on, and Ignition is ready to go, and then I can hyperlink to open Ignition into a new tab, which I've done here.

32:35
Benson Hougland: So I'm gonna go ahead and log in. I'm gonna go right to Config and log in to Ignition Edge running on the EPIC. So I'll do so here. Ignition has its own set of credentials, so I'll go ahead and enter those in. And the first thing we want to understand is, well, how did Ignition Edge get all the tag data from my control program and its IO? Well, that couldn't be simpler. The beauty of Ignition is it is an OPC client, and the EPIC is an OPC UA server. So we just made a quick connection there between Ignition Edge and the OPC server built into the EPIC. So pretty slick. So what do the tags look like? So I'm gonna switch now over to the Ignition Edge designer, and we know it's Edge because there's a little green icon up here that usually gives us an indication that we're working an Edge.

33:18
Benson Hougland: And indeed, here are all my tags; they're all coming in. Now I could have just done OPC tags here, but because we wanted to be able to scale quickly, I've created a UDT definition that will work for all stores. So as I open that up, I can see all my tags, all my subsystems are set up there, and then I instantiate my tags in the main tag folder, and that's where I get to put in some parameters that allow me to model that data and also connect to any C store that might be in the system. So pretty slick there that I can get all that in. As you can see, I'm already starting to build out the first part of my UNS infrastructure. Also, you'll notice here I've been able to develop a local GUI. So now what for anybody that's in the C store? They'll have a UI that's available right on the front of the EPIC, and it's also available to say an HDMI monitor that's connected to the EPIC or from any mobile device that may be on the local network at the store.

34:17
Benson Hougland: So this provides, again, some local UI, really easy to build that there. This is a very simple example of what I've done there. Okay. So I'm gonna go back to the Ignition Edge gateway, and here I am again; it's green, so I know I'm there. And the other part of this system of course, is how are we gonna get the data, all that modeled data that we've modeled at the edge and move that up to the MQTT broker. So as you can see, I do not have it enabled yet, so at this moment I'm not sending any data up to FactoryStack, and obviously I'll be able to connect it here once I'm ready to show you how that's done. So a couple of things. I did set up a transmitter, and of course, that's set up here. I've got actually a couple of transmitters, but the one that we're working with here is the webinar.

35:05
Benson Hougland: And as you can see, I've got my Sparkplug IDs that identify what this system is. But here's an important one. I have a history store, and that means that when I'm connecting to the broker, should that connection fail for whatever reason, whether the system goes down or communications are failed, we are storing the data from the C store locally. And then as soon as connections resume, I'm able to send all that historical data up. This is important because it means I'll never miss all of my C store operational data. So very slick there as well. Alright, now the next thing I'm gonna do is I'm gonna come back over here to the servers, and under servers is really where's the place I'm sending this data and there it is. That is the the endpoint up on FactoryStack where I'll be sending all my data.

35:54
Benson Hougland: And as you can see, I'm not currently connected, so we're gonna go ahead and make that connection. But before I do, I'm gonna jump over to the Ignition designer up on FactoryStack and again, orange icon. Now I'm up on the full Ignition server, and I can see on MQTT Engine, I've got two stores, the two stores we're working with today, El Dorado and El Dorado Hills and Folsom. But now what I wanna do is actually turn on the Temecula store and start seeing if we can populate that data. So I'm gonna switch back to the edge gateway, go to general, and simply enable. Now, at this moment, we're gonna collect all that model data that we have at the edge, and we're gonna start burping it up all through MQTT into the Ignition running on FactoryStack. This is a live demo, so we're hoping to see that information pop in here real quick.

36:49 Benson Hougland: And there it is. Right here. We now have Temecula. So now we have all the C store data from those modeled UDTs landing in FactoryStack in Ignition. Pretty slick. So let's see if it showed up on the map. So we're gonna come back to the map, and boom, there it is. I've got my C store; I can see my HVAC, kind of like what Kevin was showing for the Folsom store. In just a few clicks, we've added a brand new store to the entire architecture. So this is really important as we start thinking about things like scale and being able to bring stores up quickly. Of course that was one of the objectives for our customer. All right. So now what I'm going to do is go back to our Perspective here. I'm sorry. We're gonna go back to the designer up on the Ignition in FactoryStack. And here we are and one of the things that we've done is we've used MQTT to move all the OT data up to this level, which is up on Ignition, and it's all in Sparkplug, and that's wonderful. We've got state, we've got store and forward, we've got everything we need. But we haven't really talked about the UNS yet.

37:54
Benson Hougland: So what I'm gonna show you now is a new tag provider we've created that does describe, in our case, our contribution to the UNS. And as Kevin showed a moment ago, I've got my stores in here; there's California, there's El Dorado Hills, and there, of course, is now Temecula. So I can start adding other data to the UNS and manage all that right here from within Ignition. And as I add more things about supplies, transactions, anything that we need to know, we can all set this up into the UNS so we can publish it back out for others to consume. And let's talk about that real quick. I'm gonna go back to my browser, and I'm gonna actually log into the Ignition gateway through the gateway interface.

38:40
Benson Hougland: And when I do so... And again it's orange, so I know I'm on the main gateway. And let's try that one more time. This is a new feature that's in MQTT Transmission running up on the main gateway and FactoryStack. And indeed it is a UNS transmitter. So we're gonna grab all the tags that we have in our new UNS tag provider, and we're gonna transmit those back to the broker. Now that's pretty cool because now we have another place in the broker where people can consume that data, and it's not necessarily Sparkplug data, and it's all follows the UNS. So just to give you an idea of what that looks like, I'm gonna open up a free MQTT explorer. You can download this for free from the internet. I've got a bunch of brokers in there.

39:26
Benson Hougland: I'm gonna scroll down to the FactoryStack one and I'm gonna connect. Now, watch this. Suddenly my UNS just shows right up here in this broker. All my stores are in there. There's Temecula, Folsom, El Dorado Hills, and in the Temecula C store, I'm gonna come down to the car wash and just take a look at the run status here. There you can see it and everything is in JSON. That means anybody with permissions can consume this data. And look at that. We just turned on the car wash, and there you can start seeing the status of the car wash as it runs through the wash, rinse, dry cycle, and then comes back to zero. So we've just democratized the data from Temecula, added a Temecula store, democratized that data for anybody to consume. And my guess is that Mark, you're probably ready to consume some of that data and do something interesting with it.

40:22
Mark French: Yes, thank you, Benson. Here let's there we go. Okay. Yeah. So because Benson has exposed that data in the UNS, I'm able to consume that from the broker and start to do some calculations with that. And for this scenario, I tried to put myself in the place of the C store or CEO. I realized that's a lot of Cs, but I thought, Hey, how can I grow this business? I focus on the car wash aspect of the business. So here at Ignition we're seeing some calculations return from SepaIQ, and I started by thinking, well, I wanna track winners and losers, so I'm going to set up a leaderboard. In so doing, I realized, hey, I needed some additional context to that raw data, and so I was able to set up some additional calculated items and supplement the raw data coming in so that I could furnish those calculations. Now with that leaderboard information, then I can drive the business decisions.

41:20
Mark French: Where do I need staffing, advertising, and so forth? Now I focus on the El Dorado Hills location a little bit more than the others, probably because that's where our office is. But I started to look at capacity for the car wash. So tracking based on a vehicle ID, which I added to the context, hey, where do we have open capacity? Again, by publishing that information back out to the UNS, other services can consume that information. Maybe we have a marketing automation platform that can target ads for that location for low usage hours where we have a lot of spare capacity. Well, as interesting as car washes are and calculations around car washes, a lot of the interesting stuff for me is behind the scenes, and I know we're really short on time, so just wanna touch on a couple things real quick.

42:13
Mark French: For many, this is probably your first time seeing the SepaIQ interface. Notice this is the backend. So Ignition fantastic front end; we're using Ignition. Democratizing that data and serving clients everywhere. SepaIQ is a back-end tool for analytics. So we have the capability to organize our incoming data; that's what these groups are. We also can organize our requests or the analysis that we publish. We can also administer our machine learning models, whether they're trained locally or externally. Of course we have connectivity, cluster management, and much more. I wanna show two things real quick. One is the API that's built in. So everything we can do on screen, we can also access from the API. And that's really exciting for me because this community is moving in that direction with 8.3 and API-administered configuration. So pretty cool for your deployment plans and expanding rapidly.

43:11
Mark French: Last thing I wanted to show real quick was the flexible nature of the structure for data. Here on the request builder screen, this is where I can lay out what kind of analysis I wish to publish and what kind of structure I want to apply to it. Obviously I'm playing kind of loose here with individual locations, but then I used a roll-up feature to aggregate data from all locations. There's a nice feature to give it a test run right here before I publish it back out over the MQTT broker, for instance. And you can see there's metadata around every execution. So I can see how long did it take, how many rows did I process. What's the metadata around the analysis? What are the results? Oh, you guys are clicking on the Folsom, the Folsom car wash.

44:00
Benson Hougland: We're sending a lot of data to you, Mark.

44:02
Mark French: Yeah, I love it. So of course this is subscribed so it's updating in real time, and my Ignition instance is also subscribed to SepaIQ to get those things right away. So we're over time on this part. So I'm gonna hand it back to Kevin 'cause he has more good stuff to show you.

44:23
Kevin McClusky: Sure. Thank you so much, Mark. Really appreciate it. We had some really good questions come in here as well that I wanted to quickly answer and then also to provide a little bit more detail on a couple of somewhat fundamental questions. And that could be from more of the group here as well. So there are a couple of questions just about UNSs in general. My screen should be popping back up here in just a second. And so one of the questions, well, a couple of the questions. "Can UNS be considered a version of Tag Taxonomy? and just wanted to make sure I'm understanding correctly. The MQTT broker is the UNS." Well, I decided to pull up the Inductive Automation definition of UNS from our, What is a Unified Namespace article, which is simply a UNS is a standardized way to organize and name data, and it contains an enterprise's structure and events in one communication interface. So the organized and naming data is a tag taxonomy. That's another way to phrase that. And so that's an important part of UNS. But the other really important part is the accessibility, the democratization of that data, making it available to the entire organization. So it's, those two pieces of the UNS I think are both very important. So it is a tag taxonomy, but I don't think you can have a properly built UNS without having a technology that properly shares it as well, makes it accessible, gives you a single source of truth.

46:00
Kevin McClusky: And the MQTT Sparkplug B structure allows for devices like the EPIC here to be able to automatically send those out, contribute to this UNS without having another server, be the owner of the whole thing. So individual pieces can contribute to this overall company UNS that is used basically everywhere. So it's a really good way to do an implementation of UNS. It's not the only way, but our customers have seen a lot of success go in that route. One other question that came in, and Benson, I think I'm gonna throw this over to you, is it says, "Hello, I think of MQTT as a good protocol on IIoT, but not as a replacement to OPC, connecting to PLCs. But recently a client is asking us to use MQTT for all of the communications. And I'm starting to test this, but it doesn't seem easy to implement MQTT in the major PLC brands to work with Ignition. Are there some ways to connect them directly or just by using an edge device? What are the options there?" So Benson, over to you for that question.

47:13
Benson Hougland: Sure. So, yeah, it's a good question. And MQTT is like OPC in that it is a communication protocol to move between two endpoints. The big difference is if you have an OPC UA server, say in a PLC, then you may use an OPC client to pull that data in. The inverse is true for MQTT. Really what you have is on a device, let's say the EPIC; it does support native MQTT, so it'll publish its data to a broker. So you can actually use them both simultaneously, and they work very well together. So in this case, it's really just a matter of, if you're on a plant floor, OPC usually works pretty well, but as you start to get into distributed environments, that's where MQTT really shines because, again, it's publishing its data on change rather than OPC, which is not necessarily scanning but subscribing to tags in an end device.

48:04
Benson Hougland: At the end of the day, it comes down to this: does the PLC that you want to use support MQTT? If it does natively, then you're in good shape, and you can use either, and it depends on the use case. If it doesn't support MQTT, then you probably will use an edge device, an edge gateway. Of course, EPIC supports that as well. In fact, we have several demonstrations where we're connecting to Allen Bradley, Siemens, and other PLCs that don't have MQTT, but we pull that data into Ignition, and then we publish that data out via MQTT. I hope that answered the question.

48:37
Kevin McClusky: Thanks, Arlen. Appreciate it. We have a few more questions that have come in that I'll circle back to in just a minute here. But I wanted to wrap up things with the demo overall, just showing a couple more pieces of everything. So somebody did ask; this is somebody who knows Ignition, of course. "I'm asked inside MQTT Engine, we have the edge nodes, we can have different things inside the edge nodes. Can I just build out a UNS this way, or am I missing something?" So inside the edge nodes, there are a couple of requirements for the way that these actually show up. So you have to have a node; you have to have the node control and the info inside here. Those edge nodes also are taking the pieces that are basically the first three pieces of the path are required to be published by the edge node.

49:28
Kevin McClusky: So you can't have an edge node. You're forced into this structure if you're not using a UNS or a separate provider for that, where you're going to have several in a row that are the first pieces instead of being able to publish into the root of a UNS, for example. So if you actually use the UNS features inside the MQTT modules, you can have a tag provider that is completely under your control, and you can publish to anywhere inside that tag provider. Of course, you don't wanna have collisions where you're publishing the same tag name to the same place, and the modules do a good job at notifying you if there are collisions so you can trace them down and take care of that. But basically you can publish two anywhere inside this UNS; you're in full control if you're using the UNS features here.

50:16
Kevin McClusky: Alright. So that said, going to jump back over, you can see that inside here there's... I can remotely start; I'll do this one more time. I can see one of these moving along there from the cameras there freezer. I could see how the fuel system's running. And these are some screens that we put together for this demo. Now, of course, since there's a UNS, this isn't the only way to do these screens. Anybody could tap into this UNS, tap into the data streams, and create their own set of screens, create their own data, create their own analytics, like Mark was just showing there. And in fact, we have some companies who did just that. So these were a couple of folks at our Ignition Community conference who came in and did demonstrations directly for, they spent a couple of days building out applications based on this same set of data.

51:10
Kevin McClusky: So this is one from a company called DMC. They're a systems integrator based in the US. This is one that was created by a company called BIJC, a systems integrator based out of the UK. You can't see it in this screenshot, but there is  animations there. The waves at the bottom move as the rent cycle goes on. And here's some information about pricing. They're pulling in historical pricing as well. They've got charts and graphs that relate to all of that. And so it all builds off of the data that is available inside this UNS. And just a quick summary of what we were able to achieve here as we're getting close to the end of our time. But the Epic stores basically inside this demo, what we were showing off and what we tried to demonstrate and hopefully did a pretty good job on, is that you could see how the application empowered the Epic C store's organization to reach the goals, able to enhance efficiency, streamline operations with making data accessible everywhere, having that data flow up to a central place, being able to analyze it across all of the different locations, and having those locations be completely dynamic, where new stores coming online immediately are discovered, available, accessible inside the applications with no additional work needed, no additional integration work needed.

52:23
Kevin McClusky: It's done once, and then it's basically reused everywhere automatically by systems that understand UNS. I'll mention a couple of other things, and we'll get to a few more questions here. So this really brings our demonstration to a close, but the webinar here wouldn't be complete without me mentioning something about the upcoming 8.3 release. We have Ignition 8.3. It's going to include a new Event Streams Module. That Event Streams Module is going to give a powerful new way to centrally manage event-driven data from a whole variety of sources. If you're developing applications inside Ignition and you're processing events, tag changes, new things that are happening that you want to process in a standardized way, changes coming from the outside, people pressing buttons, even if you want to manage that, central event streams are a great way to process those events from a centralized management standpoint. It's also great for connecting up to event streaming technologies to be a stream processor for things like Kafka streams that are coming from the outside.

53:27
Kevin McClusky: So a lot of companies IT departments are using systems like Kafka or other message queuing systems that event streams can be great at processing and interacting with and sending data back over those. Perspective Module has a new drawing editor, new form, generator form, component, offline mode, which is going to be really nice for companies that do paper forms and don't have connectivity remotely. You'll be able to use Perspective; you can create applications that do that. And then there's an additional set of functionality inside the gateway. We've made everything restful. Everything is web services, so you can program against the gateway. We've moved everything to disk, so it's very friendly for source control tools. We added a whole slew of IT-focused features, security-focused features, and there's a lot more there. Feel free to check out our websites and our previous webinars as well, if you want some more details there. Some information from ICC this last year. Lots and lots of great features. We're extremely excited about 8.3.

54:31
Kevin McClusky: And if you liked the things that you saw today, I'd also wanna say, let's team up and advance Digital Transformation. Advance this concept of Unified Namespace. We're very excited about it. Hope you are too. You can learn more about all of the partners, our website there under Partner Solutions. We'll send out links after the webinar here too. And if you've never tried Ignition, please feel free; go ahead and download it. And it only takes three minutes to get going there. We do have Inductive University, which is a free online e-training site, and we have a set of international distributors that are spread across the world covering multiple different territories. If you're not on this list, then you're not covered by a distributor; you can always reach out to us directly.

55:15
Kevin McClusky: We are happy to do business with anyone from practically anywhere in the world. So please feel free to reach out to us too. If you are in one of these territories, we have an international distribution manager, Yegor, and then obviously we have contact information for each one. And if you'd like to speak with our sales representatives, we're available here. Numbers on the screen, and I'll answer just a couple of questions here. I wanna respect everybody's time, but I also wanna get to couple that are the more burning ones. So Mark, one for you. "How is SepaIQ and 4IR solutions? It differentiates each other by an application perspective." So I think the question is, how is SepaIQ versus just a standard Ignition application. What are the benefits of SepaIQ there?

56:12
Mark French: Yeah, thanks. So I tried to outline that in the slides. So take a look again, but afterwards, but again, we're adding value by adding a contextualization layer, adding additional calculations, easily exposing machine learning models within those calculations. That's a drag-and-drop prediction feature, essentially. So there's a lot more going on there. So I think when we're talking about Digital Transformation, we have to consider where the data is going to be used above these layers, and most of the cost of those higher-level analytics, like AI tools, are absorbed in data contextualization and cleaning. So, we're trying to fit into that space to provide value there. But while we're doing that we also want to provide those analytics directly back down to Ignition, facing our operations audience, mainly in the HMI/SCADA layer. So that's just a couple of things, but if you wanna do a deep dive on it, feel free to reach out to us. I'd love to jump on a call with you, and you can ping your Inductive account rep pictured here to do that. James, did you want to mention anything on that because you were mentioned?

57:32
James Burnand: Yeah, I was just gonna say as partners with Sepasoft we've actually been working with them as well. We're not quite there yet, but to offer SepaIQ as part of a deployable part of a FactoryStack instance as well. So we're not quite there yet, but that is part of our discussions that we're having and plans that we have.

57:53
Kevin McClusky: Fantastic. I'll do a couple more questions, just super rapid-fire, and then we'll wrap it up here. One is, "Is there a way to quickly publish all the PLC tags to MQTT?" The quick answer is yes; if you have a PLC and you pull all those tags into Ignition, the MQTT Transmission Module does just that. So pointed at that folder, it will publish everything to MQTT. You can choose the format. By default it'll be Sparkplug, but you can do JSON formats if you want to there as well. That's up to you. So very quick, very easy. Any tag that supports browsing it's just click, drag, and drop, and a little configuration, and you're off and running. And the other one is, "Can UDT definitions from the cloud gateway propagate to the Ignition Edge instances?" And the quick answer is yes, you need to set it up appropriately to do that.

58:41
Kevin McClusky: And generally speaking, what you'd end up doing is a remote tag provider over Ignition's gateway network. So if you're using Ignition at the edge and you're using Ignition centrally, then Ignition supports the idea of remote configuration, remote adding or updating of UDTs. You can use the Enterprise Administration Module to manage those things. If you're using systems outside of Ignition, then your mileage may vary. But if you're staying inside the Ignition ecosystem, then sure, you can certainly move those definitions around and that can be a good way to manage things. Alright. Thank you so much, everyone, for sticking with us. Thank you so much for our guest presenters here. It's been really fun to talk about. Probably gotten the idea that we're very passionate about this. So everybody in the audience, please feel free to reach out to us at any point. We do have another webinar next month, and we will be back with that. Until then, stay connected to us on social media, keep up with our newsfeed. We have blogs, articles, case studies, all of that content going out on a regular basis. We really appreciate you being here today, and if we didn't get to your question, please reach out we will be happy to answer it. So thank you so much. Take care, everyone.