Exploring The Relationship between Edge Computing and IoT

Oluebube Princess Egbuna Transcript

Intro - 00: 00:01: Welcome to Beyond the Screen: An IONOS Podcast, where we share insights and tips to help you scale your business's online presence. Hosting genuine conversations with the best in the web and IT industry, and exploring how the IONOS brand can help professionals and customers with their hosting and cloud issues. We're your hosts, Joe Nash and Liz Moy.

Joe - 00: 00:22: Welcome to another episode of Beyond the Screen: An IONOS Podcast. We have an extremely special guest today, Oluebube Princess Egbuna, affectionately known as Bube. She is an seasoned software engineer who has gracefully navigated her path into the fascinating world of cloud native technology. Bube's extraordinary commitment to open source communities, including the Open Source Community Africa, Women Who Code, and Facebook Developer Circles has proven instrumental in shaping her extensive understanding of the ever evolving tech landscape. Her professional experience from her time at Spectro Cloud to her various leadership roles has equipped her with invaluable insights. With a key eye on emerging trends and a deep understanding of technological intricacies, Oluebube is ad-divicate for Edge Computing, a technology that's revolutionizing how businesses operate and interact with the digital world. Bube, welcome to the show.

Oluebube - 00: 01:06: Thank you.

Joe - 00: 01:07: Thank you for joining me today. So we obviously covered a whole bunch of things in that intro. And there's a lot of different terminology and things that Alissa's might not be familiar with to dive into, but I want to start with a bit about you and your journey. So we mentioned that you work in the Cloud Native Space. Can you tell us a bit about your journey from being a software engineer to entering the cloud?

Oluebube - 00: 01:28: So thank you for the question. A lot of people have dramatic experiences that make the choices that they've made, but for me, it's been nothing dramatic. I've been really intentional about being a good software engineer and being at the heart of the tech community. So throughout my career as a software engineer, I've also been at the heart of the community. At a point in time, I've led engineering teams, I've worked on open source projects, I've contributed on source, and last year I had the opportunity to do something challenging because it's been the same thing over and over again. So I wanted to do something really challenging, something new. I wanted to feel like I was adding value, and that was why I jumped on the opportunity to join Spectro Cloud. I wanted to add value, I wanted to do something new. So it's nothing dramatic for me. It's just those intentional decisions about being at the heart of the tech community and also still contributing to open source at the same time. So at Spectro Cloud, I started working on an open source project called Kairos. A lot of Kairos, almost everything about Kairos is edge. That was how I dabbled into edge computing, and that was it for me. It was nothing really dramatic.

Joe - 00: 02:46: Right, it just came into it for that project. So the edge is, it's the concept of Edge Computing has been kicking around for a little while now, but at least in my perception, that definition has changed to be refined over the time. So for our listeners, would you mind telling us what edge computing means to that project and to you?

Oluebube - 00: 02:59: So we are in the era of data, right? Everything around us emits data that can be used for something really meaningful. So if we try to generate insights from the data that everything around us emits, we can have improved lifestyles. So traditionally, this data was processed in, like we just send this data to a central server and then that tries to process the data and then give us insights based on the data. But nowadays, we want to process this data closer to where they are generated and that's what edge computing. So edge computing is basically the practice of processing and analyzing data close to the source of the data. And this is important for a number of reasons. Faster processing of data reduces latency because you have to send this data to a central server like there's a lot of latency. And there are some things, there are some decisions that need to be made on the fly and this is not going to be possible if we have to send our server to be processed. And there's also the case of autonomous driving, for example.

Joe - 00: 04:07: Where the latency, the round trip is a lot.

Oluebube - 00: 04:09: Exactly. If you have to send those data to somebody, there can be collisions and a lot of all those accidents. So that is what Edge Computing is really about processing data close to where it's generated.

Joe - 00: 04:22: Interesting. So you mentioned the centralized servers and I guess like we talk a lot about the cloud in the show, obviously being an owners podcast. What does the compute environment look like for that? So obviously you don't like, you know, wear these, you mentioned autonomous vehicles, obviously an autonomous vehicle has a processor on it and there's not like, it's not being sent to a data center nearby. Is it all happening on device? Is that the computing setup?

Oluebube - 00: 04:43: So that's what it sounds like, but then basically computing resources such as servers and edge devices are deployed closer to data sources. So it doesn't necessarily have to be just a device. Servers can also be closer to where data is generated and that's what it is. It doesn't have to be just one device.

Joe - 00: 05:05: So you mentioned autonomous driving. So I think that's a really good motivating example. You can really see how the latency comes in the industries, like is Edge Computing valuable for.

Oluebube - 00: 05:15: So there's healthcare, right? We have a lot of wearable devices and also need to make a lot of decisions on patient data that's generated. And we also need to do some remote patient monitoring. So we find that Edge Computing is really important in those industries. There's also retail. So customer behavior changes, right? We are humans, we change over time, like purchasing patterns change, based on our mood and all those things. We need to process all those data in real time. If we're able to process that data in real time, we can be able to react to, respond to customer reactions in real time as well. So retail is another industry that benefits a lot from Edge Computing, Healthcare, Autonomous driving. And there's a lot, there's even a case where we saw a company using Drones to pick Fruits from their remote farms. Like there's a lot of things that you can do with Edge Computing.

Joe - 00: 06:16: It's fascinating to hear the retail one. I feel like there's been some version of that retail use-case for many years for many different technologies. When IoT and Bluetooth first became a thing, there was all this talk of beacons coming in and doing that kind of stuff. It's fascinating to hear it come back around with Edge Computing. So that's obviously a very different computing paradigm. We talk a lot about the cloud transition and people still moving their businesses into the cloud. What are some of the challenges that businesses are going to face in trying to integrate with Edge Computing or find a use case for Edge Computing?

Oluebube - 00: 06:47: So I don't think every company needs to move to the edge. It's not a one size fits all kind of use-case. So I think that one of the things that businesses need to do is to understand the problem that they are trying to solve. And they try to understand the problem in terms of the data volume that they are generating, in terms of latency requirements, in terms of processing needs. And this is going to guide their decision towards moving to the edge or not. And not everything needs to be on the edge. I think that businesses maintain their current infrastructure and they try to make maybe customer experience or how they deliver solutions to customers when they try to improve it on the edge. I think that's a good use-case as opposed to ditching their entire infrastructure and then trying to the edge. So I think that businesses need to understand how edge applies to them, how it applies to their solution, the solution they are trying to provide to customers and then how to adopt Edge Computing.

Joe - 00: 07:52: Now I'm curious, like, do you see, I feel like every technology has like a famous example, like a famous pattern where people think they need it, but they actually don't. Are there any cases in Edge Computing where you've seen people like move over to edge computing for a given use case where actually that didn't really need Edge Computing at all?

Oluebube - 00: 08:08: So in my professional experience, I've not seen that yet, because I think maybe the reason I've not seen that is I've not exactly worked really close with customers, you know, but I think there will be cases where people have realized, oh, we didn't exactly need this. And I think people need to engage solutions architects when they want to make those kind of decisions, because it could cost a lot.

Joe - 00: 08:33: Right, yeah, absolutely. So a lot of the classes of devices, you know, you mentioned like Drones and the sensors and the stores and that kind of thing. I guess they all come under the category of IoT. So is Edge Computing like very much an IoT-related phenomenon? Is it primarily the industry that this is serving?

Oluebube - 00: 08:51: So with IoT, they are closely related concepts, but they are not the same. With Edge Computing, you don't exactly need to be on the Internet. So if you think of it as IoT, you can as well say that maybe an ATM falls in that category, but it doesn't because an ATM is connected to the Internet. Those edge devices, for example, are not connected to the Internet and will still happen. So that's the difference between IoT and Edge Computing. So Edge Computing actually refers to that decentralized processing analysis of the data closer to the data source. And on the other hand, IoT is just a network of interconnected devices, sensors and objects.

Joe - 00: 09:40: Yeah, that makes a lot of sense. It's definitely an angle I hadn't thought of it before that, of course, if the computing is happening locally, it doesn't need to always have that persistent connection. And that makes perfect sense, you know, the car example you started with if a car goes into a tunnel.

Oluebube - 00: 09:50: Exactly. You know, you can be in a parking lot, for example, and they need to make payments. And if you make payments, the door needs to open for you, right? If that, imagine that you're standing in front of the door after making payments, because the data has been sent to a central server to be processed, then you just stand in front of the door waiting for a response, you know? So with Edge Computing, right? The cameras in the parking lot have already gotten, they have the data, right? Okay, you make the payments, the payments were successful, and then the door just opens for you. So the process happens locally. I mean, I think we've got some part of this. I'm trying to go, I mean, it's fine.

Joe - 00: 10:30: That makes a lot of sense. That's really interesting. That's definitely an aspect of that I hadn't appreciated. So I guess the other area where, you know, you mentioned the medical industry and patient data and, you know, wanting to do that close to, to know where that happens. Um, what is the security like story for Edge Computing? Like I imagine stuff being processed on site also reduces the amount of sensitive data that you're sending back and forth from distant places. Is that a component of the benefits here?

Oluebube - 00: 10:56: So some aspect of security is improved and then other aspect of security becomes a problem. And I'm going to explain. So you move data away from central servers and then try to process it locally. What's some level of data breach that you've prevented just by doing that? At the same time, because this edge devices are closer to customers, it's easy to tamper with the devices. So it's a given big thing. There's an advantage somewhere and then there's a disadvantage somewhere. However, there are a lot of solutions around this. I mentioned that I work on Kairos, for example. Kairos provides a level of security by making sure that the operating system is immutable, which means that you cannot override the data. You cannot change everything at the operating system. And even the data volume that is mounted, if for any reason you remove maybe a hard drive from an edge device or something, you can't decrypt the data on it, except to take it back to the device that you've removed it from. It's something called TPM encryption. So there are a lot of solutions around this edge security thing. It's actually a huge concern because once you move these things closer to users and customers, like it's just easy to tamper.

Joe - 00: 12:23: Yes, you have new attack vectors, right? No one's getting into a data center to take your hard drive.

Oluebube - 00: 12:27: Yes, although you've been able to avoid some data breaches that have to do with sending data to central servers, you've also created another problem, which is making it easy to tamper with data that's close to the source.

Joe - 00: 12:42: Yeah, that makes sense. So to flip around to the economic side, you know, talking about those trade-offs, how, what are we talking about in terms of like, you know, if a company is starting to deploy edge, what does that look like cost-wise for them? This is a cost-effective solution versus other alternatives.

Oluebube - 00: 12:57: Yeah. So these things are not set in stone, right? So having to process data at the edge is less expensive than having to transmit all this data. Data is generated in terabytes and gigabytes and all the bytes, like the era of data. So imagine having to move all this data to a central server to process it and then generate insights from it and send back to the user. It can be costly. And cost is not just maybe financial cost. It can cost a life. Imagine that patient data, like remote patient data to a central server to process it. And a decision needs to be made in a matter of seconds. It can cost businesses money as well. So I think there's that part where the cost of processing is reduced. However, having to set up Edge Computing, like having to set up everything that has to do with Edge Computing is also very expensive. a lot of hardware in the beginning, you need to invest in hardware in the beginning. So that's not set in stone. But then the long-term benefits, depending on your use-case can be really huge.

Joe - 00: 14:12: Yeah, it would be interesting to see the trait, because even if you're doing a centralized computing situation or more of an IoT situation, you still have hardware that sends that data back, even if it's not as sophisticated to do the processing itself. So I'd be interested in like what the cost per board compares to, but that's probably a very esoteric question. But yeah, I love it when guests say it depends. That's when we know we're getting to the good stuff. When there's no clear answers, that's like when you know that you've got to the real meat of the issue.

Oluebube - 00: 14:40: That's why I said you need to actually engage a solutions architect because you need to know the advantages and the disadvantages for you. You need to know what is going to cost you and what it is not going to cost you in a cost-mix days. I can't remember the term. It's something for gone alternatives. The cost of gone alternatives, you know. I mean, you need to factor in all those things and know what works best for you and the long-term benefits of the solution that you're adopting.

Joe - 00: 15:08: Sure, yeah, absolutely. So I guess turning over to the technologies involved, you mentioned Kairos already. Tell us a little bit about the project. I think I got the gist that it's an operating system for Edge Computing, is that correct?

Oluebube - 00: 15:18: Yeah, so it's both an operating system and a factory. So when I say that, it means that you don't need to do away with your own operating system. Kairos can transform your own operating system to an immutable operating system on the edge, right? You can use it as an operating system or not. Something, if you're using something already, I mean, all these other operating systems, Linux-based operating systems, they already provide like, they already do a good job and you don't want to really do away with the wheel. So Kairos builds on top of that.

Joe - 00: 15:52: So you've been working in the space, you've been playing with Kairos. Have you got any examples of interesting or complicated Edge Computing projects that you've worked on or that you've seen interesting takeaways from?

Oluebube - 00: 16:04: So like the example I gave initially, we saw a company that was using Drones to pick their faruits from farms. That was really interesting because I've never seen that as a use case. Like in my widest imagination, I wouldn't think that somebody is going to do that. But I mean, I think people use this technologies to their own advantage in whatsoever way they want to use. And that's the beauty of open source, right? You can adopt technologies, make it into what you want to use it for and, you know, use it. So I think that's one of the most beautiful use-cases that I've seen. I mean, I was mind-blown when I saw that. Of course, there are people using it in the healthcare industry, there are people using it in telecoms and all those other industries, but I never imagined that someone will use it to pick from the farm.

Joe- 00: 16:56: That's fascinating. And also very useful, you know, we hear a lot of issues at the moment with, especially where I'm from with getting fruit out of fields. That sounds like a really good deployment. A consequence of certain ill-advised decisions my country of birth has made, shall we say, there's now not enough workers to work the fields. And so, which is entirely their own fault. Yeah, that is an interesting side effect.

Oluebube - 00: 17:20: I mean, you can learn more about Kairos and try to...

Joe - 00: 17:23: Yeah, absolutely. Yeah, I'd love to. So actually that was that leads me on to a question which I was going to ask, which is, you know, you ended up working in Edge Computing, you know, as you said, not necessarily as a result of a big dramatic decision that just iteratively happened. For folks who are listening to this who are inspired and wants to get into Edge Computing, like what's the best way for them to learn more?

Oluebube - 00: 17:42: So for me, when I joined Spectro Cloud, I didn't exactly know that I was going to be working with Edge Computing, right? I was going to be doing just Kubernetes, but I doubled into Kubernetes on the edge. So I had to do my own research, learn more about Edge Computing, and then see how both of them marry each other, Kubernetes and Edge Computing, you know. It was a lot of work for me. And a lot of people expect that somebody's going to hold them by the hand and try to explain things and try to, I mean, you have to do a lot of work, trying to learn anything, any new concepts and new logic. So, I mean, my advice would be to do your research, see if it's something that you'd like to do, because you get into something and you're like, ah, okay. I didn't know it was that complicated, you know. Just do your own research, see if it's something that you're really interested in. And if you have questions, the tech community is a very welcoming community. Like you can always ask anybody questions. That's one amazing thing in tech. Like you can look out for people that are into Edge Computing and you can ask them questions. I mean, I'm on LinkedIn, you can send me my questions. Ask a question or try to Google it.

Joe - 00: 18:59: Awesome. Perfect. Well, that brings us to the ender's days episode. Bube, thank you so much. This has been... I've definitely learned a lot. Thank you.

Oluebube - 00: 19:06: Thank you so much. It was good being here.

Outro - 00: 19:12: Beyond the Screen: An IONOS podcast. To find out more about IONOS and how we're the go-to source for cutting edge solutions and web development, visit ionos.com and then make sure to search for IONOS in Apple podcasts, Spotify and Google podcasts, or anywhere else podcasts are found. Don't forget to click subscribe so you don't miss any future episodes. On behalf of the team here at Ionis, thanks for listening.