Christina Cardoza: Hello, and welcome to the IoT Chat, where we explore the latest developments in the Internet of Things. I’m your host, Christina Cardoza, Editorial Director of insight.tech, and today we’re going to be talking about the state of edge computing with Martin Garner from CCS Insight and Dan Rodriguez from Intel. But before we jump into the conversation, let’s get to know our guests a bit more.
Martin, welcome back to the show. For those audience members who haven’t listened to any of the other great episodes you’ve been on, please tell us more about yourself and what you do at CCS Insight.
Martin Garner: Well, thank you, Christina. So, I’m Head of IoT Research at CCS Insight, and we’re one of the leading analyst firms in the tech sector. And my focus is mostly on the industrial use of IoT across quite a range of sectors and technologies and things. And I’ve worked with Intel and Dan for quite a long time, including doing a report on edge computing, which is an accompaniment to this podcast.
Christina Cardoza: Yeah, absolutely. Excited to dig more into that report that you guys did. But before we get into it, Dan, please tell us more about yourself and what you do at Intel.
Dan Rodriguez: So, first it’s truly great to be here with both you, Christina, and Martin. Really excited about this conversation and talking about the future of edge computing. And in my role at Intel I am the VP and General Manager of the Network and Edge Solutions Group. And I’ve been at Intel for over 25 years, and during that time I spent most of my time working in both a network and edge space.
And in that 25 years, I will say the bulk of that time was within the telecommunications industry, but I also spent some time in other industry sectors. And I will say that it’s been very exciting for me to see many different industry transformations through the years and have the opportunity to truly participate and help lead the shift to NFE in telecommunications. And with that we’re starting to see additional shifts to more of a software-defined infrastructure in areas like manufacturing. And I know that we’re going to spend a bit of time today talking about that.
Christina Cardoza: Yeah, absolutely. And it’s great that you guys have such an in-depth knowledge and background in edge computing, since that’s exactly what the conversation is going to be around today. I think by now edge computing is pretty well known in the industry for being able to bring computation and data closer to where it’s generated, which allows for the real-time insights, high performance, and low latency that the businesses really need to succeed in today’s modern world.
So, we know what it is, but companies and organizations don’t really always know how to get there. So that’s exactly where I want to start off this conversation today. And Dan, I’ll start with you—just, with your vast knowledge in this space, 25 years at Intel, where do you see edge computing today? What are the trends and challenges you’re seeing, and what has been the uptake in its adoption recently?
Dan Rodriguez: Yeah, I mean, first off, we are truly seeing how both edge compute, as well as use of AI, is really driving an incredible amount of change across all sorts of industries and really fueling digital transformation. And when you just take a step back and you think about it, digital transformation is really all around us as companies are looking to truly automate their infrastructure to improve everything from operational efficiencies to enabling new operating models, but also provide them new monetization opportunities.
So when you think through this, it’s really about how companies are looking to save money or manage their TCO, but also make money of course. And with the advent of AI plus the advent of 5G, I believe those two things will only accelerate this trend. And if you think about just one industry for a second—manufacturing—we’re already seeing customers start their journey on AI. And where they start their journey, it shouldn’t be too surprising, because obviously when you’re running a manufacturing plant you want to take—you got to take—measured risks.
So when they’re thinking about AI to start, they’re doing simple things like utilizing it for supply chain management by maybe having autonomous robots help them stock or pull inventory. But I will say they’re quickly advancing, and they’re looking at how to utilize computer vision and use AI to assist with things like defect detection to help them with their overall product-quality assurance.
Christina Cardoza: Yeah, it’s amazing to talk about digital transformation. We’ve been talking about it for quite a bit, but it keeps changing. How you digitally transform, like you said, it’s a journey, and first it started with the cloud and AI and now with edge computing. So it’s interesting to really learn how companies are really taking hold of their digital transformations, and the new and exciting technologies they can always leverage.
Martin, you mentioned in your intro that CCS Insight just did a report on the state of edge computing, and of course that’s available on insight.tech—we’ll link that out for any of the listeners that want to learn more about that. But I’m wondering, from that report, what you’ve learned about edge computing: where we are, what the challenges are, so to speak.
Martin Garner: Well, yeah, thank you, Christina. I think the first thing I’d say is that, as Dan has already said, there’s a huge amount of it already out existing across all industries, including quite a lot that you might not even think of as edge computing. Things like industrial controllers and what have you are increasingly inside what we now think of as edge computing. And that highlights a couple of things about the whole edge computing space.
So, one is that it’s very broad: it runs from sensors all the way out to local-area data centers. It’s also quite deep: it goes from the infrastructure at the bottom up through the networking, through applications to AI—which Dan has already mentioned and we’ll come back to. And because of those two things, it’s quite a complicated area. There are lots and lots of technology areas, lots of individual technologies within that, and quite a lot of change on the supply side.
Most of those are good changes, making it easier to use and more manageable. So there’s quite a lot of progress there. In adoption terms, we think there are three big drivers. One is IoT. It has always been one of the big drivers; it still is. And what we’re finding there is that people are generating such high volumes of new data that they need to analyze on this—analyze and kind of deal with this in near real time using analytics, machine learning, or AI.
Recently though, telecoms guys have become very interested as a supplier into edge computing with multi-access edge computing and private networks. And, lastly, the economic climate—we’re all in it at the moment. Many companies are kind of reviewing their cloud spend, and that is a bit of a spur to do more with edge computing, because although they’re reviewing their cloud spend they’re still generating more data; they still want to do more with it and edge computing helps with that.
Christina Cardoza: Great. I want to dig into something that you both mentioned: how edge computing—it’s really spanning across all these different industries. Every business can take advantage of edge computing to be more successful in their operations and their business. So, Dan, I’m wondering if you can talk a little bit deeper into what those opportunities are for businesses across industries.
Dan Rodriguez: Absolutely, Christina. But maybe before I dive in to specific examples in industries, I’d like to just come back to what I mentioned earlier around making money and saving money. Because I will say, as a general manager, that’s what I think a lot about when I drive my own business, but also as I approach customers. That’s how I think about those two angles when I help them solve different problems or different challenges.
So, first, when you think about how to save money—companies, they want to have more control. They want to find ways to optimize their operations, their costs, their data. And we think about this current environment we’re in, we’re all—we’re seeing lots of macroeconomic challenges. It’s very volatile out there. You’re seeing supply chain challenges, you’re seeing unstable energy production, as well as there’s just challenges and sometimes labor force shortages today. So, lots of opportunity here.
And then when companies think about making money, of course edge AI can help here too. And you think about computer vision for a second here—it can provide all sorts of valuable insights and help improve the overall customer’s experience, as well as help ensure that stores can do many different great things, including even helping them with their merchandising strategy.
So let’s talk a little bit more about retail. And when you think about retail, to save costs—one of the biggest costs that retailers have is theft. And believe it or not, it’s a $500 billion-a-year problem. And through the use of computer vision with AI you can help attack this problem by utilizing techniques and technology to help you prevent theft at the front of the store—so at the checkout area—the middle of the store where you sometimes you get in-aisle shoplifting, or even in the back of the store where sometimes you have theft in warehousing and distribution centers.
And then when you think about how retailers make money, they can utilize AI in all sorts of new and interesting ways. First up, they can—AI can help them with shopping experience and driving overall more sales. It can also help with insights to provide the effectiveness and provide feedback on different merchandising-display strategies. It can quickly identify when there’s out-of-stock items on store shelves, and it can also just help keep stores more clean. So sometimes it’s very simple things that can really lead to better results for retailers.
And then one more example that I love to hit, just kind of quickly. So, first, when you think about manufacturing, which I did mention earlier, it is going through a massive transformation. And when you think about the massive transformation that manufacturing is going through, it’s really looking at the types of infrastructure that gets deployed. And then, generally speaking, they’re moving away from what I would call fixed-function appliances—or an appliance that’s doing one thing very, very well—to more software-defined systems that are easier to manage, upgrade, as well as to control different elements on a manufacturing floor.
And through this process you’re seeing these diverse kinds of manufacturing processes get streamlined onto fewer and fewer software-defined platforms, which of course increases the overall efficiency and reduces the infrastructure’s complexity. And with that, once you have this software-defined infrastructure in place, then you start combining with the use of robots, with sensing, with 5G and AI. And then you can do all sorts of magic across a factory floor to help you with everything from inventory management to defect detection. So, truly a ton of opportunity out there across many different vertical markets in edge computing.
Christina Cardoza: Yeah. And I think you hit on the biggest benefits that businesses are really trying to get, which is to save money, make money, have better control, better optimization, better operations. So I think we hit on—we know why manufacturers or why all industries want to move towards this, but, like you mentioned, manufacturing is going through a massive transformation, probably one of the biggest transformations out of all the other industries and one of the hardest, because manufacturers have it a little bit more difficult with their infrastructure in place, and they can’t always have downtime: they can’t make these changes and then stop the entire factory, because then that’s going to stop the whole entire production lines.
So, Martin, I’m curious to hear from you since you focus a lot in this industrial space and we’re talking about manufacturing, let’s dig a little bit deeper into that and what challenges do you see manufacturers face as they try to transform their operations with this edge computing approach?
Martin Garner: Sure. And I don’t mean—in saying this I don’t mean to knock the opportunities at all—they are huge—but honestly there are a few challenges. Now, some of those are faced by everybody who’s trying to use edge computing. The first one is scale. Edge computing is one of those technologies where it’s quite easy to get started and do a few bits and pieces, but as soon as you scale it up in the way that some of the manufacturers need to do, then it all becomes a bit more tricky. So now the larger players are going to have thousands of computers on tens of sites across, say, seven or eight geographic regions, and they have to keep all of that working, updated, and secure, and synchronized as if it was a single system to make sure that they’re getting what they need out of it.
Now, linked to that, with a large estate of edge computing you end up with a really big distributed-computing system, and then you have things like synchronization of clock signals, synchronization of machines, synchronization of data posts into databases. And all of those can be a bit tricky, and not everybody is just good at them to start with. On top of all of that we have different types of data going through the system, a different mix of application software, some cloud, some multi-cloud, some local data. All of that needs a proper architecture—that architectural complexity is also there.
But there are a couple of other challenges which are maybe specific to manufacturing and production industries. So, one is real-time working. This is a special set of demands that, by and large, IT doesn’t have. So, in the manufacturing industries you often have feedback loops which are measured in microseconds. You have to get the feedback there in a very, very short time. You have chemical mixes measured in parts per million. And so timeliness and accuracy are incredibly important here. And what’s really important is that that’s a system-level thing; it’s not just one component, it’s the whole system has to cope with that.
And then, Dan has already touched on this, the sort of robustness of the system. Many factories work three shifts per day, nonstop, 364 days a year. And an unplanned stoppage is a really expensive thing—millions of dollars per day in many cases. And so all of the computing has to support that. And so now we’re talking about things like system redundancy, hot standby, automatic failover, so that if something goes wrong the system doesn’t stop. Now what that means is that you have to be able to do software patches and security upgrades live without interrupting or rebooting the systems at all.
It also means that if you need to expand the hardware—say you want to do a new AI algorithm and test it out on the production line and so on, you’ve got to be able to put that in without stopping the production line. So hardware and software need to be self-configuring and cannot break other things down. And, again, those are constraints that IT doesn’t have. So in the industrial area we need to get used to those as things we have to work with.
Christina Cardoza: Yeah, absolutely. And a little bit of a downer, because manufacturers have all of this momentum to change and want to be successful, and then they hit all of these roadblocks that you mentioned, Martin, on their way to this digital transformation or change. So, since you mentioned all of these various problems and challenges that manufacturers are facing, I’m wondering what can they do about it? What have you seen? How have you seen manufacturers successfully approach this?
Martin Garner: Well, yes, the challenges I listed made it seem a little bit gloomy, but it’s not. So, first thing we would recommend, and have seen people doing, is don’t build your own infrastructure. It is tempting when you start designing a system to kind of design your own networking to go with it and so on. But it’s too slow, it’s too much resource, too expensive over time, and it’s a specialist area. And a bit like loading apps onto a mobile phone, there are good ways to do it and there are bad ways to do it, and everybody needs to think in a similar sort of way with edge computing.
Second thing is to design the system around modern IT and cloud-computing practices. Edge computing needs to work well with cloud services, and that should be almost seamless across the two. In practice a lot of edge computing architectures are very similar to the cloud at many levels of the stack. The main difference is that the machines are smaller and a bit more constrained. And of course in edge computing there are lots of good technology frameworks to choose from. So that what that means is most of the customer-design work can focus at the application level, and that’s where it should be; they shouldn’t be designing all of the stuff underneath.
Now the third one—in the operations-technology world typically we see that equipment and software lifetimes are 10 to 20 years in factories. We think with edge computing it’s sensible to plan for shorter lives, 5 to 10 years or so. And the reason is that the data volumes are going up and up and up and up, and the more data you get, the more you want to do with it, and the more you can do with it. So you’re going to need more AI, more edge computing capacity, and you’re going to have to expand what you have quite quickly.
And then the very last one—historically a lot of customers prefer open-source software to avoid vendor locking, but they actually then see it as something that they have to support as a specialist thing in-house. Now actually the supply side on that is changing a lot, and there is good commercial support for open-source systems, and so we can start to see those alongside the commercial-cloud offerings. So there are a number of significant changes in the way that people can approach this which make it more likely they’ll get a successful outcome as they build out edge computing.
Christina Cardoza: So, a lot of challenges but not impossible, just having a strategic approach to all of this. Dan, I’m wondering, from your perspective, you mentioned manufacturers now taking advantage of things like the Industrial Internet of Things, AI, data analytics to prepare their factories for the future and keep up with these changes. So, wondering how you’re seeing manufacturers approach or adopt this type of technology. Any lessons learned?
Dan Rodriguez: Yeah. So I think—look, I’m going to kind of frame it in terms of the journey that they’re on, and I think that the first part of this journey, it’s really what I mentioned earlier: it’s really the movement away from single-function devices or fixed-function appliances to more of a software-defined infrastructure. And once you move to a software-defined infrastructure then you can consolidate multiple workloads on fewer and fewer devices, and that can have a huge impact on both flexibility and agility, as well as just overall TCO.
So when you step back and you think about what that looks like in practice, historically you may have seen three or four different devices, all with their own computer systems—they all have the independent computer systems. With software-defined systems you can centralize those workloads into a single device and still meet the needs of time-sensitive applications. Now, Martin did mention the phones, so I can’t resist this analogy: can you imagine if you had a specific phone for each application that you had? That would be kind of difficult for you to manage. So you could think about the same thing on a factory floor. Yes, they know how to manage this today, but think about how much more easy it would be, how much complexity we would reduce, if you could actually load more applications onto fewer software-defined infrastructures.
So with that, let’s take kind of a simple example. Let’s think about adding machine vision into a robotic platform. Historically, again, you would just add a dedicated computer for this. However, if you think about where companies are going—and really the future is that you’ll have servers, and servers will host most of, many of these software workloads, and then you’ll be able to provide automated updates in a much more controlled fashion to make it much more easy and efficient to operate and maintain all these different robotic platforms that are going to be across your factory floor. And then when you think about this future, and you think about having this consolidated server infrastructure with this more software-defined layer there, you can also layer on all sorts of new capabilities—everything from quality control, defect detection, to situational and awareness.
So when you think through this, and you think about really being a manufacturer, and you really obviously have to step back and you think about that’s how to solve that business outcome that you’re seeking. And when manufacturers go through this process they really think about what processes they have, what technology they have access to, and what technology they can deploy. And then thinking through the data sets that you can use to capture information to analyze and really help make the best possible decisions overall. Whether, again, it’s that quality assurance or it’s just managing inventory.
Christina Cardoza: Yeah. And building on that phone example, if you had to have a phone for each application that you have obviously that would limit you on the number of applications that you would have, or in the manufacturing space where we’re talking about, that would limit you in the changes and transformations that you are able to make. So, good to hear the different changes and approaches that are enabling manufacturers to really be a part of this movement.
Martin, I’m wondering if you can provide us with some examples. We’ve talked about the approaches, but do you have any specific examples you can share with us on exactly how these industries or businesses have used these approaches and what the outcomes were?
Martin Garner: Yes. So, we’ve found a couple in doing the research for the piece that’s available as a download, found a couple that were really quite instructive. One came from a large oil and gas company, and I was quite surprised but I saw the logic. They run three completely separate networks. So, they have their OT network, where all the machines link together. They have an analytics network for doing all of the detailed work on the data. And they have their IT network for everything else. And they, for security reasons, they insist on an air gap between those. And so we can’t just connect edge computing up to all of them and make good sense of the whole lot across the piece. They have to be separate in this company. And that clearly adds quite a layer of complexity to how you tackle things; it doesn’t prevent you doing anything, you just have to think about how to do it in a way that suits them.
The other one, the other example we found highlighted the scale issue. It actually wasn’t manufacturing; it was a hospital, and they were installing a mesh network to keep track of ventilators and other key equipment and gather information from sensors. And they did a trial with battery-powered nodes and sensors and the five-year battery life from the supply. That all sounded great and they did the trial. That all went well and they loved it. But they realized that as they scaled it up on a very large university hospital they would have thousands of devices with batteries to monitor. They would always be changing batteries somewhere, and they’d risk a lawsuit if they hadn’t done it, which is really dangerous.
So they asked the supplier to produce mains-powered versions instead, because you can’t always get power and connectivity to all the places and the sensors that you need. So the lessons that came out of that for me were the suppliers have to design in the scale that they’re going to face and the security to support the computing from the start. And the customers need to think big at the design phase. They need to work out: “Well, how big could this be? And what would work in that scenario? And then let’s start from that premise and take it on from there.” But I think, as Dan mentioned, it is a journey and you learn a lot as you go through.
Christina Cardoza: Yeah, absolutely. And one thing that comes to mind as we’re talking about this is we’re talking about these big changes in these organizations, and some of the examples you just mentioned—manufacturing space or hospital —these are large bodies of businesses that it takes a lot to be able to, like you mentioned, scale and just do this successfully.
And a theme that we always have at insight.tech—we talk to a lot of different partners and companies—is it seems that nobody is doing it alone: you can’t go on this journey without help from others. And, Dan, I know Intel has a whole ecosystem of Intel Partner Alliance members that can help—bring some expertise, or help manufacturers and other businesses along their journey to be successful and to adopt edge computing. So I’m wondering if you can talk a little bit about the importance of partnerships and how you guys leverage them.
Dan Rodriguez: Absolutely. And when you think about the overall partnerships, I do think about creating an overall ecosystem that is truly diverse and utilizes open and standard platforms. And I think that that was really vital in the—you think about the transformation that happened in the telecommunications industry and the movement to NFE, but it’s also going to be incredibly important as we drive towards this change in manufacturing as well as other industries. So I’ll leave you with maybe one example here.
We’re involved in something called the Open Process Automation Forum, and it’s truly a great place to really democratize technology advancements in manufacturing. And as an example of this we’ve been working with companies like Schneider, Exxon, Dell, and VMware, who came together with us to pull together and drive new and open industry solutions and deliver these open technologies in a field trial to really showcase how you can utilize next-generation automation techniques across manufacturing.
And I will say it does take forums like this, like OPAF, to truly make something like this happen and also really scale it across the industry. So when I think about the ecosystem, it’s incredibly important for the overall health of the market to have a very vibrant ecosystem that utilizes standards and open-based technology so the community can not only have a lot of vendor choice but you’re also increasing that overall innovation spiral and advancements in technologies to really solve those business outcomes that manufacturers are seeking.
Martin Garner: If I could just add one thing onto that. I think I mentioned earlier that edge computing is broad, deep, and complicated. And from what we can tell, very few customers can take on all of that. Very few suppliers can take it on either because they tend to specialize in certain things. And so actually most of the systems we’re talking about will need to be designed with three to five players involved. And I think that’s the expectation we should all bring to this, that it’s going to be a team effort all round.
Christina Cardoza: Yeah, that’s a great point, Martin: it definitely takes a team to make all of these changes happen, and especially the theme that you guys mentioned is to not only make it happen but to be able to scale it. We’ve mentioned things like utilizing standards and not being locked into things.
So, Martin, I’m wondering if you can expand a little bit on that. As businesses start to make these changes, as they want to scale and look towards the future, how can they make sure that the investments or the changes that they’re making today really future-proof their efforts so that they’re able to be flexible as their approaches and their needs change in the future?
Martin Garner: Yeah, and I think we probably covered some of this a little bit already. So the core of it is to build a system that’s as flexible as possible. And in practice that means using commercial hardware and doing as much as you possibly can with the software on top. Anything that’s sort of locked between the software and the hardware is much less flexible and will—the age of that will tell.
The other thing is to emulate things that really work well in other tech domains. And we talked about mobile phones, but I think the app-store concept as a way of being able to download software—one of the great benefits for manufacturers is the flexibility that edge computing can bring. If you can just download some software, reconfigure your machines, and start producing something different fairly quickly, that is a huge benefit that they’ve just never had before. Now obviously an app store or a phone are a bit of a simplification when it comes to factories, but I think the concept is good and we need to kind of embrace that sort of idea.
Christina Cardoza: Yeah. And as we’re talking about future-proofing and looking towards what’s coming next, Dan, what do you see coming next? How will this space continue to evolve over the next couple of years, and how will the role of edge computing change in industrial environments as we move forward?
Dan Rodriguez: Absolutely. And I would think about this—as we all know, things need to go through phases and it is truly a journey. So the first phase, as I mentioned, it’s really about that migration towards software-defined infrastructure, with workload consolidation supporting multiple applications on fewer and fewer servers or devices. And once that’s established then all of a sudden you can do all sorts of cool things with AI and inferencing to really help you across your factory floor, improve the overall output of your production, but also do statistical analysis to even help with the health and preventative maintenance of all sorts of equipment on your factory floor.
And then I will say obviously generative AI is all the buzz across many industries today, and it will, over time, be incorporated in this strategy as well. And I’ll say that it’s going to be super exciting to see all the gains in production, reduction in defects, and also the use of new simulation and modeling techniques of that factory in the future.
And then, Christina, you did mention healthcare earlier today, so I do want to just maybe say a couple quick things on that as well. And when you think about just the broad role of AI and edge in healthcare, it’s going to do many different things including even helping physicians—really assist them—and helping to improve patient diagnosis as well as treatment. Really enabling them to have more timely detection as well as decision-making.
In addition to that, across healthcare we are starting to see the broad use of distributed computing to enable all sorts of AI-based use cases, including drug discovery and diagnostic tools to truly power connected digital hospitals and labs, and also support real-time data analytics across that entire kind of hospital footprint. So, lots of exciting times across manufacturing, healthcare, and then obviously I talked a little bit about retail earlier.
Christina Cardoza: Yeah, absolutely. Martin, is there anything you wanted to add or that you found from the State of Edge Computing report that you guys did? How the role of edge computing is going to change or evolve, not only in manufacturing but as we get into some of these other industries as well, like retail and healthcare.
Martin Garner: Yeah, sure. Thank you, Christina. There were a couple of things that came out that—not so big right now, but you can kind of see them coming, and enough people that we talked to sort of waved a small flag to say, “Keep an eye on this one, because it’s coming.”
First one is around the fact that manufacturing processes for the company—those are mission critical, and any unplanned downtime, as we said, is really, really expensive. So there’s a key question about how do you learn from things that have gone wrong and ensure that those mistakes are not repeated. Now, the aircraft and the processing industries have always been quite good at this. They have this concept called functional safety, and their aim is to make systems more and more resilient when things go wrong by making sure that the failure modes are understood and mitigated, and progressively they build in new scenarios so that they can kind of cope better under fault conditions. That to us looks like an important area for more general use across manufacturing, although just today it’s not so large.
And another one is linked to the industrial robustness that we mentioned, and I really like this one. If an application can run on one machine and automatically switch over to another one if there’s a failure, well then you get a question about, well, which is the best machine for it to run on normally? What’s the optimum setup for these thousands of computers and all the applications? What’s the right way to do it? And as soon as you think about that, you realize that optimum could mean fastest, it could mean the lowest latency, it could mean the highest uptime, cheapest on capital costs, cheapest on operating.
There are lots of different parameters that you could optimize for here, but really it’s all about optimizing the system in different ways for different things going on. And customers won’t want that to be a complex process; they’d like it to be automatic if possible. We haven’t found anybody who’s actively exploring this yet, but we do expect it to become a thing fairly soon in edge computing, and to see some sort of software tools come into the market that allow you to improve the setup of your edge computing estate over the next few years.
Christina Cardoza: That sounds great, and we’ve been talking for a little bit now. Such a large conversation—I always feel like we only scratch the surface in some of these conversations. I know there’s still a lot more to learn in this space and a lot more for edge computing to go, but unfortunately we are running out of time. So, before we go, I just want to throw it back to each of you—any final thoughts or key takeaways you want to leave our listeners with today? Dan, I’ll start with you.
Dan Rodriguez: Yeah, no—I appreciate the conversation today, and I will say there are a few final thoughts that really come to mind for me. First, edge computing is really fundamentally changing nearly every industry, from retail to manufacturing to education to health as well as transportation. And second, when you combine edge computing alongside AI and 5G it’s driving a lot of transformation. It’s allowing IT, OT, and CT to truly converge and really accelerate digital transformation, and this truly creating a massive opportunity and really the opportunities are endless. Everything from precision agriculture to robots that sense and cities that intelligently coordinate across vehicles, people, and roads.
And third, I do strongly believe that industry collaboration and open ecosystem are fundamental and key to all of this. As Martin mentioned, it is going to be a team sport, and you’re going to need multiple players to drive these solutions and implement them in a way that’s easy for customers to consume the technology and easy for them to be able to scale the technology.
And with that, Intel, along with the rest of the industry, is truly investing to drive this unified ecosystem that really understands the pain points across many different industries and helps them solve them, and, again, in a way that’s easy to deploy and scale. So I look forward to continuing to working with all sorts of customers and partners on this journey, and of course working with Martin as he analyzes this journey and provides guidance to all of us in the industry.
Christina Cardoza: Absolutely. And Martin, any final thoughts or key takeaways?
Martin Garner: Yeah, thank you, Christina. And Dan, thank you; that was a great job of describing the vision. And it’s a little bit of an analyst cliché to say, “Oh yes, but it’s complicated,” but it actually is complicated, and I think for many companies who are involved, whether in manufacturing or other sectors, that vision—they can kind of see it and they get it, but it feels quite a long way away for them. And so, from our point of view, from the research we’ve done, we think it’s quite key for customers to get started, to do something, even if it’s quite small. And when you do that, pick out a few carefully chosen partners and work with them.
We think at the start you should be fairly ambitious in how you think about all of it—what scale could this all get to—and you won’t get there all in one go. You’ll probably find though that it’s not the technology that’s the limiting factor in how far or fast you can go; it’s probably the organization, and that often is quite a limiting factor—whether that’s budget or other organizational factors. So you will need to invest at least as much time and effort into bringing the organization along with you as you do in working out what technology to use and how much of it and where and so on. But that is the journey, I think. But I don’t think anymore that the technology is the limiting factor, and that plays to Dan’s vision, I think, really quite nicely.
Christina Cardoza: Yeah, I can’t wait to see how, like you guys mentioned, edge computing with a combination of things like AI, computer vision, and IoT not only continues to improve business operations but changes people’s lives. So I encourage all of you to keep up with Intel, follow them, see how they can help—their ecosystem and their partnerships—how they can help you along this journey, as well as all the technological advancements that come out to, like we said, make things a little bit uncomplicated.
And also take a look at the report from CCS Insight on the state of edge computing on insight.tech. There’s a lot more information than what we just covered today on how industries can start tackling this and the technologies involved there. So, with that, I just want to thank you both again for joining the conversation; it’s been very insightful, informative. And I want to thank our listeners for tuning in also. Until next time, this has been the IoT chat.
The preceding transcript is provided to ensure accessibility and is intended to accurately capture an informal conversation. The transcript may contain improper uses of trademarked terms and as such should not be used for any other purposes. For more information, please see the Intel® trademark information.
This transcript was edited by Erin Noble, copy editor.