Explore the Digital Oilfield

Rising to Business Challenges by Improving Data Access

62 min video  /  49 minute read View slides

About this Webinar

Even though the oil and gas industry is facing serious technological, economic, and cultural challenges, companies in this space can remain strong and competitive if they are willing to shift their thinking and adopt new solutions.

In this webinar, experts from Inductive Automation (creators of Ignition, the first universal industrial application platform with unlimited potential), Streamline Group (a builder of efficient technology solutions for oil and gas and other industries), and Cirrus Link Solutions (a leader in IIoT, MQTT, and SCADA solutions) discuss a comprehensive approach to innovation, scalability, and efficiency that boosts operations and saves money.

This discussion is based around a real-world implementation that has resulted in improved network performance, the creation of a field-data distribution hub, and reduced spending on upgrades to the core SCADA environment, along with other huge benefits.

What you will learn:

  • Current trends having the greatest impact on oil and gas
  • How to accelerate innovation and efficiency in your organization
  • How to slash operational costs
  • How to improve executive decision-making through real-time data access throughout the organization
  • How to develop and execute a comprehensive approach to achieve efficiency
  • Why decoupling pipeline SCADA systems from devices that control equipment helps accelerate innovation

Webinar Transcript

Don: Good morning, everyone, and welcome to today's webinar, Explore the Digital Oilfield. Appreciate your time this morning. My name is Don Pearson, I'm Chief Strategy Officer with Inductive Automation, and I'll be moderating today's webinar. Quick look at the agenda. We'll start with an introduction of Inductive Automation and the Ignition software, I'll introduce our panelists, I'll give a brief presentation on the state of the industry today. Then we'll have a roundtable discussion about challenges in industry and about an implementation of our combined solution at a major mid-stream oil company. We'll hold an audience Q&A at the end as we always do.

Don: A little background on Inductive Automation. It was founded in 2003. From the very beginning, we've been an independent company with no outside investors. We're pleased, very pleased, actually, that enterprises around the world have chosen Ignition for their HMI, SCADA, MES, and IIoT needs. It's been installed now in over 100 countries. We have an integrated program that has over 1400 integrators used by lots of... Hundreds of oil and gas companies around the world, as well as in practically every other industry that you could imagine. Since 2010 when we launched the Ignition platform, we've really continued to experience explosive growth with an average annual growth rate that actually exceeds 60%. As far as info on the company, if you're interested, if you're new to Inductive Automation, certainly just go to our website to the About Us section and you can get as much information on the company and the principles and the industries we serve as you're interested.

Don: As I said, it's used by thousands of companies, including 41 of Fortune 100 and about a quarter of Fortune 500. Almost every industry, not just oil and gas, but the water and wastewater, food and beverage, government, transportation, packaging, and honestly, the list goes on. With all the oil and gas companies around the world using ignition, we've done several oil and gas case studies over the years. Some of the benefits that users and integrators have told us about are remote access, real-time data from thousands of remote devices, database accessibility, unlimited tag history, and operator efficiency. Those kinds of things are the benefits that people are realizing in oil and gas. Also, when one looks at the non-technology side of things, certainly there's been praise for the low cost, the hassle-free licensing model, and really strong service and support. So if you like, you can read about our oil and gas case studies at inductiveautomation.com. Just click on Solutions, and then click on Case Studies.

Don: Just a brief on Ignition, it's the first database-centric cross-platform, web-deployed industrial application platform for HMI, SCADA, and IIoT. Six quick reasons why it's web-based deployment. Certainly its unlimited licensing model is important, it gives you unlimited tags, clients, connections, devices, projects, unlimited concurrent designers. It's really... It puts you at the driver's seat in terms of innovation with this platform. It offers security and stability, the module architecture makes it easily expandable, it's set up for rapid development and deployment and offers real-time control and monitoring.

Don: We have a really good group of knowledgeable panel members today. Gregory Tink is the managing owner of the Streamline Group, a builder of efficient technology solutions for oil and gas and other industries. He provides oversight for Streamline Group, focusing on oil and gas technology, data management and analytics, field networking services, and niche product and service development. That's the quick bio I read for you, Greg, but why don't you give us a little bit more on yourself?

Gregory: Oh, thanks, Don. I don't know... I won't dominate the whole thing by talking about myself, obviously. I left Encana in 2010 with the intention of starting up a company that looked at OT and IT, sort of the field experience and how the enterprise data requirements coming from the field were growing, and in 2011, I started up with a couple of partners Streamline, and we have been evolving how companies are gathering data and turning it into information from the field for the enterprise systems ever since. And as we began our journey, we were looking for some strategic partners that could help round things up. We're focused on solving business problems, and we try to be technology-agnostic as much as possible, but obviously we look for favored partners, and we found Inductive Automation and Cirrus Link along the ways and realized that we had some common perspective on where the future was going with regards to industrial control environments. And now we're currently working with a number of oil and gas companies, a utility company, and a few others, with Ignition, but also with the implementation of IoT and the strategy around it, specifically with middleware as well.

Don: Thanks, Greg, and thanks for your time today too. I'm also pleased to welcome Arlen Nipper. Arlen has more than 37 years of SCADA experience, including SCADA system implementations for Fortune 100 oil and gas companies. He is the President and CTO of Cirrus Link Solutions and the co-inventor of MQTT, which you'll hear more about today. But Arlen, tell us a little bit more about your background and what brings you to this area today.

Arlen: Thanks, Don. Yeah, for the purpose of this discussion specifically I have been in the oil and gas SCADA sector for almost 38 years now, came out of Oklahoma State and got the opportunity to work for about eight years for Copal Oil Company. And the irony is that the SCADA system that I put in for Koch in 1979, looks very similar to the conventional legacy SCADA systems that I see installed today. So over my career, going from Koch to co-founding a couple of technology companies, we were really entrenched in the infrastructure side of SCADA, I.e., protocol converters, protocol converting protocol, A to B and protocol B to C, and it seemed to be a never-ending story. At about the time that AT&T got deregulated, VSAT technology became a viable alternative to dedicated phone lines at the time, and we had to look at better ways of getting data back over VSAT, because you couldn't afford to poll.

Arlen: So at that time, I had the great opportunity on a project, on a real-time SCADA system project for Phillips 66 pipeline at the time, to work with Andy Stanford-Clark from IBM. And what we basically did was took... We kind of mashed together two worlds that had never been joined before. IBM came from the standpoint of having belts and suspender messaging middleware, and at the time, our comm control systems came from the standpoint of having to make poll response data systems more efficient and faster. And from that, what evolved out of that 18 years ago was this transport technology called MQTT. Now, having said that, MQTT was relatively hidden in the background initially with IBM, but lately it's emerged as one of the dominant, originally dominant IoT protocols. And what we've been able to do, and, Don, you've heard me say this before, is that we invented MQTT 18 years ago, but it took 18 years to find a platform like Ignition to actually leverage it. So very excited to be on this discussion today.

Don: Arlen, thanks. Appreciate your time also. Travis Cox is also on our panel today, he's Co-director of Sales, Engineering at Inductive Automation. Travis started here in 2003, and he's played very important, actually leadership roles in the development and establishment of our training and support services, our design services before taking over his current role in sales engineering. So, Travis, a little more about your background.

Travis: Yeah, so I started here in 2003, so I've been with Inductive Automation for 13 years now. I came from an IT background and found working with an integration company in this industry, and so I got to see the legacy architectures and the conservative approach, if you will, to CA systems. And really, my goal here has been to leverage and bring OT and IT together with Ignition, and that's something we've been doing from day one. And today being part of sales engineering, my team is there to help give a high-level technical expertise to customers to help with their architecture and set-ups of these systems. So I'm happy to be here today as well.

Don: Great, thanks. Thanks to all you gentlemen for joining today. Just a few comments in terms of where the industry is heading. A lot of disruption going on, been going on for a while, when you look at the state of the industry, you see it's caused by a variety of things, certainly IoT, big data, the cloud, mobile technology, social technologies, and I think the list goes on, but those are some pretty major ones. But they all have something in common, and the thing that they have in common is data. They all increase access to data in order to connect people, organizations and systems. With that increased what they call the three Vs, the volume, the variety and the velocity of data, it really starts changing everything else. And companies are really searching to try and get to that fourth V or the value of data, and they're investing in data as never before. And I think that's why we see big industry-leading companies really changing the way they've done business for many years. Smart Industry magazine has called this transition, the ongoing digitization and datafication of industry.

Don: Looking broadly at the state of the oil and gas industry, we see a variety of factors that all support the need for change. Some of them are listed here, aging infrastructure, global competition, the price of oil, technology trends driving change, expensive DCS system upgrades are just no longer a viable option. Overall, just an open standards-based secure and interoperable control system, facilitating innovation, age integration, it's present in other industries, so why not oil and gas? Well, along that line, I think some big companies are making some big choices. Exxon-Mobil, we heard them presenting at the ARC Conference in February this year. They're pushing to innovate their processes through open architectures and interoperable solutions. And one of the reasons that Exxon chose Lockheed Martin as their integrator is aerospace is way ahead in this area, and they wanted to develop a next generation open and secure automation system that would apply to process industries. According to Automation World Exxon-Mobil has been vocal about challenging traditional automation technologies and practices, really in order to shorten project cycles, decrease capital investments, increase efficiencies. And they've been asking automation suppliers, don't just bring us incremental changes, but let's get innovative and let's look at things differently.

Don: I think another point is the rise of platforms. Arlen mentioned in the introduction, looking for a platform and eventually finding that Ignition would do what he needed to do, but in the industrial space, there's real change, I think. Automation World's Editor-in-Chief, Dave Greenfield, wrote in an article that there's a confluence of trends that appear to indicate a move by both automation suppliers and end users toward an automation platform approach. So sort of going away from just getting a discrete device or a software application that does one thing or a specific operation, let's get the platform. I think the Advisory Group of ARC also predicted suppliers of all kinds will tout the power of their platforms in 2016. I would certainly say at the shows that we've been at this year, we've seen that to be the case. And almost everybody there, platform, platform, IoT, IoT. Travis also mentioned, actually back to the beginning, Ignition has really been in, and he's had a major role in this company, on sort of bringing OT and IT together, but probably the biggest effect of these disruptive technologies is this ongoing convergence between OT and IT. And I think it's worth maybe giving it just a little bit deeper look.

Don: On the IT side of the ledger, a range of software, hardware, telecommunications, other services used primarily traditionally at the management level for information processing, generate data for the enterprise or facilitate communication. The IT personnel, very good with SQL, Java, Python, etcetera. OT is used for machinery, other physical equipment, monitoring control systems, and is built from the plant floor up. You look at the OT and controls personnel, they're great with PLC programming, the HMI/SCADA side. So really, these two groups have two different skill sets and they lived in separate worlds pretty much for a long time. And then things changed. IT and OT functioned as those separate worlds.

Don: But in this GE blog, there's a quote that puts it very well, "IT worked from the top down deploying and maintaining data-driven infrastructure, largely to the management side of business. OT built from the ground up, starting with machinery, equipment, and assets and moving up to monitoring and control systems. For a long time, these two divisions kept to their own turf and found their own effective solutions to problems. Then came smart machines, big data and the industrial internet, and the worlds of IT and OT suddenly collided." Greg, on that point, just because of your background, you had experience in dealing with this wall between IT and OT, can you make some comments on what you think keep... Some things keep these two sides separated, from your experience?

Gregory: Yeah, I have a lot of experience with this. Obviously, as I mentioned, Streamline wants to be seen as a company that has one foot on both sides of the firewall, one foot in IT and one foot in OT. I personally have a lot of experience operationally as well. And from the beginning when SCADA systems were first put in place in the field, and there's still a lot of reasons why there's a struggle around this. One of them is, while network improvements have been amazing over the last few years, there's still some gaps out there. Devices are still remote and there's still a sense of isolation in a lot of cases, and there still are automation islands. Those are breaking down, and mobility is taking over, but there's still that problem with the network and it's evolving as we go along. One of the biggest reasons why there's still a little bit of a gap and an issue between OT and IT is what I like to call the philosophical hangover. And it's really supported by the fact that the worlds were so isolated in the past, and it's taken some time for companies to really be able to figure out how to support bringing those worlds together. And we'd like to help evolve that and bring those worlds together, but a lot of times there's a resistance to change on both sides because they're wary of each other and they've sort of gone down the road and been burned a few times before.

Don: Sure.

Gregory: And then there's the technical issues as well, like proprietary protocols and devices, and there's not a lot of usage on those devices with protocols like MQTT. So there's still a literal gap between those two worlds.

Don: Sure. Thanks, Greg. And I know as we work towards the solution to the technical side, that's not gonna change the reality of the mindset shift that has to occur. The organizations like ARC, certainly, and trade publications have commented, but the Gartner group made their comment too with this quote, and the bottom line on this particular quote is that the end state sought by organizations is really this integration between IT and OT. And it's moving forward, but there are gonna be those challenges, some of which Greg just pointed out, as one goes forward. There are just too many benefits to overlook it though, just some of them listed here, but the list is quite longer than this. More and better information to make decisions, reducing costs, lowering risk, optimizing business processes, faster development integration and standardization of communication and controls, so that it's possible to get that data into the enterprise where it can have an effect.

Don: The bottom line is it's gonna require a different architecture that de-couples protocols from applications by using message-oriented middleware, or MOM, 'cause MOM provides more efficient information distribution, increased scalability and shorter development times. For example, just the ability to discover data and automatically generate tags by simply subscribing from a client, saves tons of development time. Publish, subscribe, replaces traditional polling methods and data goes into infrastructure instead of going directly into applications. That way that data can become an asset for the entire enterprise, not just data going to the benefit of one application.

Don: Ignition actually is... I know it was mentioned by both Greg and by Arlen, it is the first universal industrial application platform, and it really does bridge the gap between the enterprise and the plant floor in the field. And it really is a facilitator. Since day one, we were database-centric and we were... We didn't call it OT/IT convergence 13 years ago, but that was the goal, and really, it's the foundation for IIoT and practically any other industrial automation application you can build. So Ignition empowers you to actualize the potential of the IIoT because of the rapid developing and deploying, the upgrading, improving, the scaling solutions, some of the things I mentioned at the outset today. The protocol that we use in our IIoT solution is MQTT, it's lightweight, it's very efficient, it's publish and subscribe protocol, and as Arlen mentioned, it's quickly becoming very well known in the standard communication protocol for IIoT. But since we happen to have the co-inventor on our webinar today, I think I would like you to make some comments about MQTT and give some details of its origins and how it's come to be adopted as an IIoT protocol. Arlen, can you do that for us?

Arlen: Sure, Don, thanks. Like I mentioned, MQTT was unique in that at the time that we developed it, we were just trying to solve a problem, and... We didn't... The cool aspect of this now is we can talk about decoupling devices from applications. When we originally started on this, it was not to do that, it was to save money in getting more data from PLCs, flow computers, RTUs, gas chromatographs, cathodic rectifiers in the field over VSAT networks, which at the time were very, very expensive. So, not to go into all of the details of MQTT, but the fact of the matter is we knew that we had to create an operational SCADA system that was as good, if not better, than a conventional poll response system. So by doing that, MQTT became very efficient. Andy and I, when we were working on it, we got to the point where we kind of said, well, every byte that we save is worth a nickel, and one of us would come in and said, "Okay, I was in the shower and I just figured out how to save 25 cents."

Arlen: But the main point of MQTT is that, to your point, Don, you're not coupling devices to applications, you're connecting devices to infrastructure so that applications that want that can come in and subscribe, so you've created a one-to-many architecture. The second thing of MQTT is that it has state, and what I mean by that is that once you have an MQTT session established and you've published all of your tag information, you're using all of the underlying advantages of TCP/IP, so now all you have to do is publish any data that changes. So you're not sitting there doing round-robin polling, polling 100 billions and... Say you poll a valve status, it's closed, I poll again, it's still closed. I polled it a thousand times, it's still closed. So those were the key components to MQTT. And then because it is simple, it's that "keep it simple, stupid" approach, is that it's been widely adopted by everyone from DIY hardware running on Raspberry Pi to mission-critical systems that have been deployed for the last 18 years.

Don: Thanks, Arlen, I really appreciate you giving us a little summary of the background there. We're gonna get into a better, further on, look at the infrastructure and this change here, but two just bullet points here. The emphasis is edge of network device is published to an MQTT server that's either on or off-premises, they're easily pushing data from thousands of devices to a central location, and now, as Arlen said, the data goes into the infrastructure, not just to one application, and it's there to be used for both industrial and business applications, can all have access to it and subscribe to the data need when they need it and how they need it.

Don: Today, our approach is just a collaboration between these two organizations, Inductive, Streamline, and Cirrus Link. And with that, I think I'd like to get into our topic here and dig in a little bit more, let's move to the panel discussion. I'll just start with an overall question, what are some of the biggest technology challenges affecting this industry? And Travis, do you wanna maybe start with some comments just from your side?

Travis: Yeah. Excuse me, I think the biggest challenge that's facing the industry is certainly the aging infrastructure, and how we can take the existing legacy systems and transition them over to a newer architecture while maintaining the existing systems so that we can easily switch over and have the benefits of this newer technology and architecture that we're gonna be talking about a lot more here today.

Don: Thanks. What other comments? Greg? Or Arlen? Just the overarching look at the biggest technology challenges?

Arlen: Well, from my standpoint, Don, it is that everyone that we go speak to about Ignition and this overall solution, they're stuck in a catch 22. So to Travis's point, it's the protocols, it's the equipment, it's the applications, and you're in a catch 22 where you can't upgrade one unless you upgrade the other, and you can't upgrade the other unless you change out your infrastructure. So having this solution and being able to put together a migration strategy I think is great.

Gregory: I think one of the...

Don: Greg, you started to make a comment, Greg. What's your thoughts?

Gregory: Yeah, sorry. One of the things that I think we need to talk about when we're talking about technology challenges is the challenge to the industry right now, around... Because there is a consistency issue, there's a lack of standards, which makes these things expensive to move down the road and change out. To Arlen's point, you need a migration strategy and you need to really look at that, but ultimately what we're seeing right now, it's not a technology challenge, but it's a challenge in how available money is to make technology changes, and that causes some change resistance around, okay, we recognize we have this issue, there's more and more of an appetite to move forward with things like middleware and then Industrial Internet of Things concepts, but we don't have the money right now. And ultimately, one of the things that's, I think, holding up and causing some hesitation is special interest groups and politics, and the... Both in the United States and Canada, there's some... An element of uncertainty around politics, and I think these types of cultural or outside influences are causing some technology challenges on their own.

Don: Sure, sure. So as we take a look at... I mentioned Exxon-Mobil's initiative. Let's take a look at how technology is progressing in this industry and the direction it may go. And certainly we're already started into this discussion, but let's take a little bit deeper look at it, because technology is progressing, there is a motivation for change, there's plenty of challenges there, no question, but there's plenty of benefits to be gained. So, Arlen, I think I'd like to start with you and your thoughts on this question.

Arlen: Well, I guess there's two fronts, there's the software side and the hardware side. Real quick, I'll start with the software side. One of the things that IBM did, that Arcom and Eurotech did when I was working with them, is that we all worked together to push as much of the MQTT software technology out into open source as we could. So I think a lot of people on this call will be aware of the Eclipse Software Foundation, which is an open foundation. They have a Paho project, that contains all of the information, the code, MQTT servers that you would need to start looking at and learning MQTT and message-oriented middleware infrastructure. From that, trends like cloud services, hybrids where you can run on-premise but have cloud availability. And then there's the fact that now we're... The devices that we're putting into the field have the intelligence that we need to make decisions out there. So again, it's funny, I talk to these younger engineers as we're going out into industry, and they're just coming out of computer science school, and they're going, "Gee, Arlen, why do we have to poll a PLC to get all of the information back just to sort through it to see what changed?" So to Greg's point, I think we're having a generation, that new digital generation that's coming up that knows the technology, and they're much more willing to apply that in some of these solutions.

Don: Arlen, I think that's great. Travis, do you wanna comment on that?

Travis: Yeah, talk about the perfect blend of OT and IT. When you look at the cloud, leveraging the cloud infrastructures, there are really great systems. As far as having unlimited storage or having applications to do machine learning and intelligence and mining that data and getting more information out of it, not only to get it using infrastructure, but also to get into more of the business parts, like ERP systems and CRMs, getting the data to these different systems, you can leverage all of them and really, really get the most out of them.

Don: Greg, you may wanna make some comments on this too, but I'm on a... And please do on this question, but I'm also gonna bridge to a next question because I'm gonna direct it at your direction, so any comments on technology progressing and also the cost-effectiveness of these solutions? When one looks at upstream oil and gas, I think I'd lean on your expertise to comment on that.

Gregory: Yeah, absolutely. I think one comment on the previous, I think we're moving away from being SCADA-centric with operation technology, and I think Arlen mentioned this as well, that we're moving towards more of an operational integration, data analytics-driven environment. The technology's there, it's a pretty exciting time right now because it's truly able to enable innovation and solve some really key business problems. And one of the key points of that, of course, is being able to do this in a cost-effective manner. And historically, the systems that we're talking about really aren't that cost-effective and they don't encourage innovation, because once they're locked in and loaded, you don't wanna change them because it costs so much to put in place.

Gregory: If someone has to reach for their wallet every time they have an idea or they want to help their business, they'll opt to keep things the same, and by putting in this technology, you can actually make some changes quickly. And cloud-based systems, like Travis mentioned... And I truly believe that cloud-based SCADA is the future for us and I think it'll prevail, the cost savings will prevail over any of the other challenges that we find. And I think the evolution of cheap devices that will sit on well sites and enable you to get data, regardless of whether SCADA is part of that, but it enables you to get data to make key decisions and turn data into information quicker, those cost-effective technologies that allow you to do all that are there right now, it's just a matter of figuring out the strategy on how to put them in place.

Don: Sure, sure. You know, Greg, I think you make a good point. We've noticed at Inductive Automation that there's an interesting silver lining to the price of oil demanding cost effectiveness and demanding efficiency and spending money very well because of the need to lower costs. But it amazes me when I think of IIoT in oil and gas and broadly is the licensing model of Ignition is perfect for that, because how are you gonna connect tens of thousands of sensors and thousands of devices? You actually need a server-based licensing, unlimited tags, unlimited clients model, or you're gonna stifle information right where it starts. So I think it's an important thing when one looks at cost-effectiveness, you look at the technology, the deployment, the effective use of bandwidth, the protocols, and the entire business model of the ecosystem you're working with.

Travis: Yeah, I wanted to mention that, certainly, Ignition, to that point, with its unlimited license model, really allows you to innovate and to get more of that... To make more sense and do more with that data, but when you look at this, obviously there's not gonna be a zero-cost solution, there's gonna be some cost to migrating to a new architecture. But thinking just about the long-term investment and the ability to... If you can get to 80% more of your data, what does that mean to your organization and to what you can do with that information that you couldn't have done before because of the traditional polling models you have? So MQTT, in all of its beauty, the biggest thing you can do is get to more of that data and to bring it into the entire enterprise, not just to the SCADA system.

Don: That's true, and I know, Arlen, you've commented on that on more than once, 80% plus of the data has been isolated out there at the end, not getting into the enterprise to be used for those business decision-making things that Travis is talking about. Any comments on that from your side, Arlen?

Arlen: No, that basically summarizes it.

Don: All right, let's move on here. What are some of the big environmental and regulatory challenges? Greg, you mentioned something about government and politics, so let's throw this question your way to start with.

Gregory: Yeah, there are a ton of government and regulatory influences and uncertainties. I think there's varied regulatory influences that are out there, and I think the challenge that companies have with managing some of the sometimes-conflicting messages... Regulatory bodies that reference each other, but reference old versions of documentation that they've put out, and keeping it coherent for themselves, and I think that's gonna be a huge challenge. There's collaboration that's being pushed, and I think companies still have issues with figuring out how to collaborate, share information, and... About improvements as well, whether it's technology improvements or just how they're managing regulations and compliance.

Gregory: There's clean energy, carbon levies, climate leadership plans, things that we're hearing on a day-to-day basis now, and there's a lack of direction and guidance coming from the policy makers on exactly what's there. But we do know they're coming, and there is a need to change. We're really at a crossroads. We're seeing more companies look to alternative energy, which is obviously influencing oil and gas. And then there's a bit of a perception that's out there that is probably very... Well, it is unfair in a lot of ways, and the industry needs to respond and do this in a time frame where there's economic uncertainty as well, which leads to a lot of challenges revolving around regulations.

Don: Sure.

Gregory: And of course... Sorry, it all comes back to one thing, it's accessing data, being able to integrate data, and in order to showcase that you're doing the right thing as well as doing those right things.

Don: Thanks. Travis?

Travis: Yeah, and Greg, you mentioned this before, and that with a new infrastructure, being able to get to the data, not only get to more of the data, but also, to get to it faster. So when traditional polling systems took 15 minutes to get information back, when you can be a second or sub-second to get your information, now we could be faster to respond to issues and when we find leaks or these other types of things that are out there that do impact what we... The regulatory challenges that are there. So I know you mentioned that, being able to get the information back fast allows us to respond quicker and to eliminate some of these potential big problems.

Gregory: I would agree, and I would go as far to say as we have to stop talking about leak detection and we have to start using technology to be able to talk about leak prevention. And I think in order to do that, you need new infrastructure and you need to get your hands on the data as quickly as possible.

Don: Yeah, that's a good point. I wanna shift to a discussion now on a specific project that was undertaken by a leading midstream company located in North America, but there were a couple of questions here, I just wanna make a sidebar quickly to answer a couple of questions here that people are asking of individual companies, whether they're working with this company or that company. We certainly respect the confidentiality of all of our customers and honor their request for that. But I would give you an opportunity, if you'd like to know, that we invite end users to come speak at our conference, and you'll see many of them speaking at the Ignition Community Conference next month here in Folsom in September, and if you even wanna go back historically and see end users who have spoken in the past, can give you a real good feel. It's all available on our website, from the past Ignition Community Conferences, so you can see what people are saying about Ignition and using it for themselves, they can say whatever they wanted to about it.

Don: So that being said, let's kinda shift over to this company that was looking to achieve many things that we're talking about here, such as the lowering cost, improving network performance, collecting more field data, making it accessible to the enterprise, being less reliant just on traditional SCADA, and breaking down that OT/IT. To do this, basically, Streamline Group and Cirrus Link got together and started building this architecture. There were a number of objectives. Arlen, can you provide some more context about the company and elaborate on their objectives for the project, which I just listed on these bullet points here?

Arlen: Sure, Don. Very quickly, I'll go through this. So Cirrus Link was doing oil and gas workshops here about three, four years ago. We did one in the Calgary area and went through MQTT architectures. We didn't have Ignition, but like I'd say, the joke that I always say in these conferences is that before I had Ignition, I was really good at drawing really good pictures on the whiteboard, and they got it. They were at the point where the legacy SCADA system and the fact that all of the applications were packed up behind it, they just needed that way to break out of the catch-22 syndrome that we were talking about before. But their largest problem, of all these bullet points, of getting more access to data and being able to do better leak detection or leak prevention, as Greg said, but the biggest payback on the project was to be able to exponentially get their data in quicker. And not only that, by using MQTT, reducing the bandwidth by about 85%, so that now, not only can they get the real-time operational data quicker, now they can afford to bring back more of that data that was being left in the field. So basically create this process control data hub where now you can start plugging your applications into it. So again, it was, line of business drove a lot of the demand for this project, better access to more data, but the primary overarching was, we need to get our data in quicker, and then from that, we'll go into the network architecture.

Don: On that point, I just wanna ask you maybe just parenthetically to make one comment, 'cause I heard these gentleman speak at our conference that we did up in Calgary with Streamline and you guys, and they had, I think, a bunch of objectives, and actually, I think maybe I'll kick it to you and you can mention that, Greg, 'cause you were very engaged with your team in doing this project. But they pointed out there were 11 objectives, and when they hit the first one, everybody was happy on the business side at that point just getting that one done. So can you talk a little bit about what benefits were achieved against the goals of the project for the company?

Gregory: Yeah, they sold the concept internally primarily on, we're gonna improve the network, and so the project was actually given a network improvement name so that it was very focused on how we can improve the network performance. And when they turned it on, the first day we went live with the first pipeline, the polling cycle, what used to be the polling cycle that took 15 minutes, they were getting data back within a 15-second range. And so the operators were swearing in the control room, it was quite a moment for everyone to see that type of impact, and it was disbelief because they had never seen that type of performance before. So what really the key is that you've gotta look at the business impacts that come out of network performance, and obviously, control room operators getting that data quicker, there's a lot they can do with it and there's benefits across the board. One of the keys that was mentioned is if we can get our data in quicker, perhaps we can have an impact on leak detection, as I mentioned before. Getting data into the system so that we can have more of a granular and timely view on what's happening. And I think these are just the first steps, to be honest, but I think that was really pivotal for them.

Don: Yeah. Greg, you're always so professional and very conservative, but I had a chance, 'cause you guys set it up, for me to be able to go to that control room north of Calgary and spend some time with that team there. And when they went from 15 minutes polling to 15 seconds... I'm not talking to these guys, they were ecstatic, they died and went to heaven. What does that do, you're right, to leak detection, to moving towards the leak prevention you're talking about? It was a pretty huge change, and that was only one of the goals they had for this project, but it clearly changes the game when you can make that kind of a shift from this kind of an architecture change deploying a newer technology. So thanks for your comments on that. Let's talk a little bit about legacy architecture, and then the current architecture. Arlen, I'm kicking this crap over to you, this graphic, and you can walk us through legacy and then we'll go to the next one and take a look at the middleware architecture.

Arlen: Okay, Don. Real quick, probably everybody on this call recognizes this architecture. And the problem with this architecture... It works, it meets the demands of what a SCADA system should do, that's not a problem. The issue is, if the SCADA system is in the center, so he is the data source. So what we've ended up doing over the years is we've taken perfectly good working off-the-shelf SCADA systems, we put them down on day one, they do what they're functionally supposed to do, and then on day two, we start getting the line of business saying, "Well, gee, Joe, I would like to have this piece of data from the flow computer. I know you don't need that for operations, but I'd like to have it in my electronic flow measurement system."

Arlen: And what ultimately happens every single time is we take perfectly good SCADA systems and then we try to make them look like message-oriented middleware, and they were never built to do that in the first place. And what happens in all cases is that so many custom things get bolted on, the system starts becoming fragile, and then it gets so brittle that nobody wants to innovate. I think Greg mentioned it before, that if a guy has a really good idea on what he could do with additional data, he doesn't even try to mention it because he knows all of the pain that would cause in adding extra polls to the SCADA system where it doesn't need the data. So we transitioned, we put together a migration strategy, so we worked... Cirrus Link worked with Streamline, worked with the customer to be able to start, keep their current polling system up and running, but in parallel, start moving the polling out to the field and publishing that data in with MQTT, and that resulted in the next diagram.

Don: There you go, middleware.

Arlen: Okay, so this is the infrastructure now, where they're using Alexis ready gate edge of network devices in the field. That is their TCPIP end point for both their cellular and their VSAT circuits, and all of the polling of the Alan Bradley PLCs, the Modbus PLCs, and the total flow flow computers is moved out to the edge. So now if I'm directly connected to RS232 or Ethernet, I can poll as fast as I want to. So the network doesn't become the limiting factor anymore. So all of the registered information from all of the devices at each of the field locations is gathered up by the edge of the network device and is published to an MQTT broker.

Arlen: Now, you'll notice in this architecture, there is no application that is sitting in the middle now to become a barrier to innovation. So all the field information is being published into the MQTT broker, and you'll see here that the SCADA system, the measurement system, the historian, all, then can plug in as subscribers to the MQTT infrastructure. So, I will mention that in this case, Ignition is not the primary SCADA system. So what's very interesting here is we have multiple SCADA systems plugged in to the same infrastructure. Ignition is being used for all of the metrics, and the monitoring, it's used to gauge how well the Alexis directors are working, how well the multiple MQTT servers are working, but their primary SCADA system is also subscribed to that data as well. So effectively, to Greg's point, we put together that migration strategy to where they never lost operational integrity, they got to the point where they could run in parallel, both polling and using MQTT, compare, do all of their point checks, side by side, and then as they got confidence in the system, one pipeline at a time, be able to migrate those over and result in a pure message-oriented middleware SCADA system.

Don: You know, Arlen, I think you bring up an extremely strong point there, and we're not talking here about an industry that's sitting around with a bunch of greenfield projects. We're talking about the world of brownfield and the need for migration strategies where you do not interfere with the ongoing operation as you migrate effectively over. It's a significant change in how one operates this kind of a transformation to the middleware architecture. Greg, any quick comments you wanna make on that?

Gregory: Well, I guess just... I think I mentioned earlier how we are moving away from a SCADA-centric architecture to something new, and a new way of thinking, and I think what we're also moving away from is an expensive and exclusive old model to a new model that's inclusive as far as other systems being able to use it, and less expensive. So I think that's key.

Don: Sure. Travis, your thoughts.

Travis: We've been getting a few questions about the edge of network devices, and if PLC supports MQTT, what the options are. I thought I'd just address that question here real quick. Arlen mentioned the Alexis ready gate product, it is an edge of a network device, it talks... It has over 200 protocols built in, talks to the PLCs directly and publishes through MQTT, whether that's in a... Through cellular or through ethernet or VSAT, it has that ability. There's also... For Ignition, there is a module for Ignition called the MQTT Transmission, which can convert OPC data to MQTT and publish that data up very easily, so you can use third party OPC servers at the edge and get data through MQTT, and there's a variety of other options that are out there for the edge of network. And so you can go and visit Cirrus Link's website, Inductive Automation's website, you can see a little more information about that. But I wanted to make sure that was addressed, because that is what really is important, to get the legacy to new. Manufacturers of hardware now, like Moxa and others are gonna start building MQTT into their devices. So they'll be plug and play when you buy new ones, but we're not necessarily buying new ones, we have a legacy to work with here as well.

Don: That's an excellent point. As we come to wrap up and go towards Q&A, I do just have a couple of questions that I wanna ask Greg and Arlen about Ignition. Why was Ignition chosen over other competitive products for this particular project? Greg, can you comment on that?

Gregory: Well, Arlen might be able to explain, I guess, in more detail why Ignition was chosen. I think, why we were looking to Ignition originally and Inductive Automation as far as our partnership goes it's flexibility, it's the technology platform, as it stood, and obviously the fact that it is a reasonably... Reasonably priced, I would say. The licensing is flexible, you can do a lot, and Inductive Automation is a great company to work with. We're really looking forward next to look at Ignition on the cloud, and we're currently in the process of moving forward with that with a couple of clients in Calgary, and I think that that's exciting going forward, and I would be hard-pressed to think of a way to do that with most of the other systems out there that we'd looked at.

Don: Thanks Greg. Arlen, I think you may have addressed this question a little bit even as you went through the architecture of where Ignition sat, but how did you use Ignition to accomplish the goals of this project, or in working with other projects too?

Arlen: Don, to be perfectly honest with you, this project was the driver of putting MQTT on Ignition. Literally, we were in roll-out mode, and like I told you, we needed to prove the system before it ever went live, and we needed the ability to simulate RTUs, PLC, massive loads, multiple levels of redundancy, multiple broker architectures over multiple network interfaces, and there was no tools on the market to be able to do that. So, that's when we were first introduced to Inductive about a year and a half ago. We got the components in and due to the flexibility, now they have complete visibility of all of their tags with all of the network infrastructure. So ironically, this was the genesis of the MQTT modules for Ignition today.

Don: Thanks, Arlen. Travis.

Travis: I know I'm a little bit biased because I work here.

Don: That's okay. I still wanna give you the microphone. Talk, be biased.

Travis: Ignition is the only platform out there that's gonna allow you to leverage 80% more data because of its licensing model. You wanna have 20 more people look at the application, whether it be on their PC or on their mobile device, the server architecture with unlimited licensing allows you to do that. Not to mention the million other reasons; ease of upgrades, not having to install any software on the client machines that have a project update automatically to monitor everything. You've got a lot of reasons and we could certainly sit down, have demonstrations. We can show you some of those in a demo.

Don: That's great, Travis. I'm just gonna take this last question here before we wrap up and go into Q&A and maybe open it up more broadly and say, just as a... Take maybe a minute each in any final comments you wanna add. Greg first, you and then Arlen. And then if, Travis, you wanna wrap up, then we'll go into some Q&A. But Greg, final thoughts to share with this audience today?

Gregory: Well I think one thought is that the companies that figure this out will have a competitive advantage. Back to your original discussion around OT/IT and how technology can help us solve these new business problems, like new tools for new problems type things as opposed to old tools for new problems, and I think that's key. There is a competitive advantage for companies to have out there. I think when we look at middleware and IoT and MQTT as it stands right now, I'm excited about the future with this because I think when you look at plants and refineries and how software is too coupled into their systems, I think using middleware will decouple that and by decoupling that, you're gonna have some future where you don't have to wait for a plant turnaround before you're upgrading your DCS system. And I think that flexibility and that ability to look at this differently will allow companies to innovate, and like Arlen mentioned, have five SCADA systems up, one in simulation, one production, and when you need to upgrade, you can move things in. It increases security that way but most importantly, I think it increases our ability to innovate and think of how we can use these tools differently.

Don: Sure, thank you. Hey, Arlen, how about your final thoughts?

Arlen: Well the final thought is we're finally getting there. Everybody's been talking about IoT, we're actually doing it. The diagram that we just walked through, we've effectively decoupled the devices. Don, Phillips 66 has been running MQTT infrastructure for 18 years. If you talk to them, the only piece of the original system put in 18 years ago is the middleware infrastructure.

Don: Thanks. Thank you very much. Let's go to... Comment on this. I'm just telling people, try Ignition if they want to. Travis, do you wanna make a comment on that?

Travis: I wanna say, first and foremost, the beauty of this architecture we're talking about is that you could actually try this today without any cost. The idea of having edge-of-network devices, there's software from Ignition that allows us to do that to get data into message-oriented middleware that we can try out, as well as getting into a system like Ignition. So you can go to the website, you can download Ignition, you can download the MQTT modules, we can do demonstrations and show you how this architecture works. All three of us, myself, Arlen, and Greg are all here to answer any questions you may have, so please, please contact us, let us know what you'd like to see. And we encourage you to go to the website and to take a look at all this.

Don: Actually, if anything else, this is an introduction to say there are some folks here willing to get on the phone, to get on a... To show a demo, to answer questions. And along that line, I just wanna emphasize, but particularly the questions here about... First off, there's a question here, "Is Inductive selling edge-of-network devices?" No, we don't sell hardware. So, Greg, we don't do that but Arlen, you've done a lot of work in this area and you've got... With Spark Plug stuff, can you maybe... We're running out of time, so make it a minute, but give us a quick look at the... What are people like Opto 22 and Moxa, and Hilscher doing, Magnetrol, with this whole new architecture?

Arlen: If I just go through 'em real quick. There's B&B Electronics, Alexis, Hilscher, Magnetrol, Moxa, Opto 22, Tyrion. These are all OEM manufacturers of end devices that are putting the MQTT and Sparkplug spec on which means these devices connect to an MQTT server, Ignition connects to that MQTT server, and now you have all of your tags and you can start designing.

Don: Thanks so much, Arlen. I'd be remiss if I didn't mention also, you can become very self-educated on Ignition platform by going to Inductive University. There's over 600 videos, one to four minutes long. You can get credentials on HMI, SCADA and MES schools, so please take advantage of it and go there. And now, we're gonna take a couple of minutes and take a look at any questions that we may be able to get to here. Thank you very much for bringing them in and we'll see what we can get. "Can someone comment on the affordability of the edge-of-network devices?"

Arlen: The price range runs from... If you go to the Cirrus Link website, we have Spark Plug running on a Raspberry Pi, so let's start at the $28 point and it goes all the way up to actual industrial equipment. So you're looking at anywhere from $350 to $2000-ish.

Don: Good, thanks.

Arlen: It all depends. Yeah.

Don: Okay next question. "Is this a one-way fee to the subscribers? How do you send back a comment to the lease for operational needs?" Travis?

Travis: So we talked about MQTT but not really in-depth in this call. MQTT is low bandwidth, it's secure, has TLS security, it's stateful, and it is bidirectional. You can write a command from the SCADA system that goes all the way down to the edge-of-network device to the PLC. So it can definitely be that way for operational needs and in fact, MQTT was built for operations at the beginning, to go back to Arlen's point.

Don: Good. And then this question, I think we answered, at least started, and there'll be more of an answer as we go. "What are the MQTT MOM products that are available?" Certainly, the companies working in that direction were mentioned by Arlen. "Then where can I buy the MQTT broker from?" Good question, Gregory. I'll tell you, we have a strategic partnership with Cirrus Link and the modules that they have developed. Certainly, there are modules for this but that's not the only broker, so I guess you can get brokers in a lot of different places, right?

Travis: Arlen, do you wanna comment on that?

Don: Arlen, answer that one.

Arlen: Yeah. Right now, Cirrus Link has the Chariot SCADA MQTT server product for on-premise and for cloud. IBM, all of IBM's message middleware supports MQTT, Red Hat AMQ supports it, Amazon Web Services, IoT has a MQTT server in the cloud, Microsoft's Azure supports MQTT. So that's the exciting thing here, guys, is that when you're talking about OT/IT convergence, most, every message-oriented middleware product supports MQTT natively today.

Don: Thanks, Arlen. I do wanna... Again, Greg, thank you for your time, Arlen, you for your time, Travis, you for your time. We appreciate the input. Echo what Travis said about our willingness to work with you and answer questions. You can certainly contact the panelist. Gregory, me and Arlen are on the upper right on this slide here. If you want anything from our account executives to coordinate and get something set up with the sales engineering team that works under Travis, then we'll... Happy to do that. These are the contact points. Contact Melanie at that number and we'll do anything we possibly can to just facilitate your own journey of understanding to see if those fit into your own strategies. With that, we're concluded for today. Thank you very much for attending.

Posted on August 18, 2016