The following transcript has been edited for length and readability. Listen to the entire discussion here on The Broadband Bunch. The Broadband Bunch is sponsored by ETI Software.
Pete Pizzutillo:
This episode of The Broadband Bunch is sponsored by ETI Software and VETRO FiberMap.
Brad Hine:
Hello everyone in broadband land, and welcome to another episode of The Broadband Bunch. I’m your host, Brad Hine. Today, we are on-site at the Gaylord Palms Conference Center in Orlando, Florida, the site of the 2023 Fiber Connect Conference, put on by the Fiber Broadband Association. With me today is Tom Mitchell, from netElastic. Tom, welcome to the bunch.
Tom Mitchell:
Thank you.
Brad Hine:
Tom, tell our audience a little bit about your title, and what you’re doing with netElastic.
Tom Mitchell:
My title is VP of Strategy and Business Development at netElastic, but I’m fundamentally a sales guy. I spend a lot of time with our customers, teaching them about how virtualized networking can improve their businesses. And that’s my life.
Brad Hine:
One that has been a lot more in demand in the last three years, since COVID started, I would imagine.
Tom Mitchell:
Yeah, this is true. Our business has grown quite a bit since the beginning of COVID. There’s been a lot of migrations from small service providers getting larger, either bulking up by investment, merger, or several other methods. And we see them moving to more important and more powerful networking technologies. That’s what we do. We provide those technologies.
Brad Hine:
Right, so it is a high-performance virtual routing solution, that delivers better scalability. Like you and I had talked about initially, when we met, that’s been a capability for years. I think a situation like COVID brought it more as a need to the forefront, but your background goes back quite a long way.
Tom Mitchell:
I started in the early eighties working for a small company in the New York area. We deployed Apple Talk Networks in some of the very first classrooms in the New York area. I’ve been in this industry for some time. On the routing side, I was with Brocade before I came to netElastic. So we saw a lot of service provider routing solutions. At netElastic, we do virtual networking. We make a software networking solution that runs on your standard servers. It costs a lot less. It’s very scalable, and it allows service providers to get into the business very cost-effectively, make precision investments, and scale up as they need to.
Brad Hine:
Cost-effective is a huge phrase we hear throughout the business. Every niche of our broadband ecosystem is trying to do that, with greater scalability with more efficient costs. Describe a little bit how you guys do that exactly.
Tom Mitchell:
What we do is, take advantage of some of the technologies that have come out of Intel and other places. We developed our routing stack, and then we used some open-source technologies. We combined those and ran them on standard x86 platforms to deliver solutions with all the functionality that customers need but at much lower price points. Intel, in particular, has done some really interesting things lately with multi-core processors, DPDK, and other technologies that accelerate network capacity and network capability.
In the past, customers needed to use merchant, silicon-based solutions because they really couldn’t get much performance out of the x86 architectures that were available at that time. Things have changed dramatically in the past three years, not only because of the pandemic but because of the evolution of technology. We’re taking advantage of that and delivering solutions that rival some of the more well-known vendors out there in the marketplace.
Brad Hine:
Right. Obviously, around the dotcom era, we’re talking about kilobytes, and then we’re talking about gigabytes. And we’re talking about terabytes now. Have you reached terabyte throughput, enabling terabit speeds through capacity for your customers in solving these challenges?
Tom Mitchell:
Recently, Intel did a test of our BNG platform on a particular Quanta server and measured it with a terabit of fabric performance. This is almost unthinkable to a lot of the legacy folks that have looked at virtual networking as nice, but not comparable to Cisco-like performance. In this particular case, we are delivering such a performance. But we’re doing it in a virtualized environment at terabit speeds.
Brad Hine:
That’s amazing, it truly is. It’s really interesting to see where we’ve come since the dotcom era, and through the last two decades, and now into post-COVID, and still growing and needing more capacity, everything that these companies are requiring. What are some of the other challenges that you’re seeing in your customer base that you guys are continuing to innovate and meet their needs?
Tom Mitchell:
One of the things that the pandemic has highlighted is the IPv4 problem. IPv4 address space is really small, and service providers, particularly the new greenfield service providers, just can’t get it. When they can get it on the open market, it’s quite expensive. When you think about 20,000 subscribers and $50 a pop for an IPv4 address, that’s a million-dollar investment. That’s pretty significant.
Brad Hine:
Just for our listeners, to unpack that a little bit, we’re just asking for an address.
Tom Mitchell:
Yeah.
Brad Hine:
That’s all.
Tom Mitchell:
Yeah, it’s just an address.
Brad Hine:
I’m sorry, continue.
Tom Mitchell:
Yeah, no worries. A CGNAT solution gives you the ability to take a small pool of IPv4 addresses and amplify them into a large pool. It saves the customers a lot of money. And since these things are very scarce, and you can’t support IPv4, they’ll tend to turn to a CGNAT solution like ours. Since it’s virtualized, and it’s running on standard x86 hardware, and doing a specific, single task, it’s a very cost-effective solution for service providers today. We’re helping the market in a big way by providing that sort of solution.
Brad Hine:
Well, it’s sure hot in this space right now. I know hearing from folks that have won RDOF, CAF II, and a lot of other grants that have even come before, obviously, the big acronym is BEAD right now, B-E-A-D. We’re all here at Fiber Connect here in August, waiting to see who’s going to be granted all this money.
Tom Mitchell:
It’s all the buzz.
Brad Hine:
Right, absolutely. How have you seen this change in even the last — 2017-18 was CAF and CAF II? Have you seen those things innovate even in the last few years to need greater capacity and greater efficiency that way?
Tom Mitchell:
Well, one of the things that I think it’s done is, that it’s allowed a lot of greenfield network operators to enter the market and provide more competition, which is beneficial to cities, subscribers, and communities. It brings jobs; it brings a lot of value to communities. We’ve seen that happen quite a bit. BEAD, of course, hasn’t really happened yet, but a lot of people are preparing for that money to flow, which we think is going to hit in a 24-25 timeframe. Those monies will be flowing as well soon, too. Everybody is gearing up to provide solutions to their customers that the pandemic really showed was a big problem.
Brad Hine:
In essence, to simplify this idea, obviously, every company must work within a network to secure and make sure they’re locking down all the employees trying to connect into a major hub, so they can get all their information, but still put up that wall of security so nobody can break through, even if they’re remote from outside at that physical organization. Have you seen more pressure on this in the government industry at this point, where people are working more remotely? Is it spread out evenly over school systems, municipal government, healthcare, and things like that? Are you seeing the equal needs all around, or are you seeing it more in one place?
Tom Mitchell:
I think from our perspective, we see it more in the rural markets.
Brad Hine:
In the rural markets, okay.
Tom Mitchell:
The rural markets have been underserved, as we all know, badly by the incumbents, which makes sense. They’re there to make a profit, and they need to find the best markets they can. But that doesn’t generally fit with what the country needs. We’re seeing a lot of rural networks grow, and a lot of growth with the service providers. We see a lot of new people getting into greenfield networks, particularly municipalities, public utility districts, co-ops, and the like which are very non-traditional in terms of providing broadband service to their customers. They own SCADA networks; they own fiber networks. Why not utilize those to serve your customers?
Brad Hine:
Yeah, that’s a great point. From working with some of these folks myself, and seeing what their needs are, they need to be educated in this also. A lot of them are coming from another utility that they’ve offered their community, maybe an electric utility or a telephone co-op that has offered a different service other than an internet service provider locally. Are you seeing a lot of education that you’re having to relay to these folks? What is the training like when you’re doing this? I know you guys are supporting that network, but are you seeing a ton of training that you must do for these folks who are now requiring these local virtual networks?
Tom Mitchell:
We really do. It’s a bit frustrating because a lot of these networks are bringing on talent, but they’re having a hard time finding talent. And there’s a scramble for talented people. Network engineering firms are great. They’ve been really helpful. We work with engineering firms like IP ArchiTechs and VantagePoint. And they provide a lot of good services to their customers that allow those broadband service providers to get smart fast.
Some of the things that we do, we provide comprehensive solutions that minimize the interconnection of boxes, so you can simplify. Especially if you’re starting in a small greenfield network, and serving 5,000 or 10,000 customers, it’s very easy to get going with a solution like we make. Then, as you grow, you’ve got the talent; you’ve got the people. And you can build out the more complex networks if you need to.
Brad Hine:
Excellent. Walk us through, if you could, a day in the life of a customer of yours before you, and then after they’ve implemented you. What are the trade-offs? What are the improvements that they’re making, that somebody who’s actually on the system is going to see?
Tom Mitchell:
As a subscriber, do you mean?
Brad Hine:
Right.
Tom Mitchell:
Well, the beauty of it is that the subscriber doesn’t really see anything so it’s transparent. They don’t see a difference. They just see beautiful broadband speed, low latency, and so on and so forth. The ways that we enable that are by providing solutions to the service providers. For example, by providing a virtualized solution on standard x86 hardware, the costs are really low. So a lot of the service providers are now distributing the functions around their network. Rather than centralizing a BNG function or a CGNAT function, they can distribute it around their network. That really helps their customers because it lowers the latency to the subscriber, keeps the cost low overall, and gives them a smaller fault domain and attack surface. So they can really manage their network better.
Now, there’s the expense, of course, of putting it in the field, and putting it in cabinets with power supplies. But in the fiber world, that stuff’s going in already anyway, so it makes sense to add a BNG, or a CGNAT appliance to that cabinet, and then provide that functionality closer to the subscriber. Now, they’ll see actual improvements in the performance of the network by the service provider being able to deploy this unique way. That’s an example of one of the things that we can do that make the user experience better.
Brad Hine:
Great. For our listeners, could you define a BNG, and then a CGNAT, so they’re getting a reference point?
Tom Mitchell:
Yeah, it’s a fairly technical thing. But a BNG, in the most abstract, way is a router with a server on top of it that does BRAS, or broadband remote access services.
Brad Hine:
Okay.
Tom Mitchell:
That’s basically a way to manage subscribers by reading the billing system, and then applying policy to the traffic on the router piece of the BNG. BNG is a broadband network gateway.
Brad Hine:
Okay.
Tom Mitchell:
It’s a very popular service provider appliance.
Brad Hine:
Gotcha. CGNAT, in correspondence to that, would be?
Tom Mitchell:
It’s carrier-grade NAT. NAT is network address translation. It’s a way of amplifying the IPv4 addresses you have in your network, so you’re not consuming too many and spending too much money. CG means carrier-grade. And again, in an abstract way, it’s a way it of doing NAT where you can NAT to your router in your home. But then you’re NATing the router itself to the service provider router. It’s double NATing, and it’s just a way to amplify the address space that a service provider can use to save a lot of money.
Brad Hine:
We love acronyms on the podcast.
Tom Mitchell:
Too many of them.
Brad Hine:
There are so many, yeah. Do you speak acronyms? I hear a lot at these shows, at Fiber Connect especially. Speaking of Fiber Connect, this is your first Fiber Connect?
Tom Mitchell:
This is the first time we’ve been here, yes.
Brad Hine:
Wow. I saw your booth earlier today and spoke to your colleagues. It’s exciting for you guys to be here. Welcome to Orlando for your first Fiber Connect. What do you think so far?
Tom Mitchell:
I think it’s a fantastic show. We’ve seen a lot of our customers; we’ve seen a lot of our prospects; and we’ve met a lot of new folks. I think this is something we’re going to be doing regularly.
Brad Hine:
Oh, I love to hear it. That’s great. Any stories of any brevity, or something a bit ironic that you could relate to our listeners that you’ve experienced maybe in the last 10 years of your career or maybe just at netElastic? Anything that you’ve found a bit ironic, or funny about working in the business?
Tom Mitchell:
I remember the one thing that was shocking, not funny. When I went to a telco in a very rural market, I expected to see a lot of copper. And they had a little tiny patch panel of copper left, and the whole rest of their breadboard was fiber. I said, “Wow, you guys have converted to fiber very quickly.”
Now, this was three years ago. And they’re like, “We can’t get this out of our network fast enough.” I was shocked that the rural carriers were moving that quickly to fiber and putting fiber to the home in very rural communities. It delighted me. It was pre-pandemic. And when the pandemic did hit, I could see that people were going to move to Fiber very quickly.
Brad Hine:
Wow, that’s a great story. More stories about how people are trying to stay ahead of the curve and service their communities with the best capacity and bandwidth that they can get in those rural areas.
Tom Mitchell:
Yeah. That’s great.
Brad Hine:
I appreciate you coming by and chatting with me today. You’re always welcome. It’ll be great to see you around the show the rest of the week. If somebody wants to get in touch with you or netElastic, how could they do that?
Tom Mitchell:
Well, thank you for having me. You can reach out to tmitchell@netelastic.com. You can find me on LinkedIn, and of course, you can go to our website at www.netelastic.com.
Brad Hine:
That’s wonderful. Tom, anything else you’d like to share with our listeners before we depart for the day? I’ll let you get back to some exhibit floor hours.
Tom Mitchell:
No, I’m glad to be here. I’m glad to meet you. I’m glad to spend this time together and look forward to many years of success for us all.
Brad Hine:
Awesome. Well, don’t be a stranger. I know we’ll see you soon.
Tom Mitchell:
Same here.
Brad Hine:
Thanks, Tom.
