The Encoder Got Faster While You Slept | How VPUs On Akamai Cloud Change The Cost And Scale Of Video Encoding

Cloud video infrastructure is getting squeezed from both sides: audiences expect better quality and more formats like AV1, while finance teams are staring down cloud bills that don’t scale with reality.

We sit down with Sarah Walter from Akamai for a one-year check-in on VPU-accelerated instances powered by NETINT, and what stands out is how quickly the conversation has moved from pilots to production. Sarah shares what she’s seeing across customers, from hyperscalers expanding internationally to broadcasters and streaming platforms looking for a practical step between fully managed services and running everything on-prem.

We dig into what’s driving adoption of VPUs for video transcoding and encoding on Akamai Cloud, including new locations in Europe like Frankfurt and London for higher availability designs. We also talk through the “bring your own software” challenge and how making Bitstreams available helps teams operate VPUs with a GUI instead of living in command lines all day.

For larger deployments, we break down new eight-card plan options and why density matters when you’re measuring cost per stream across live streaming, VOD, premium content, and user-generated workloads.

Operationally, we get specific about the parts nobody advertises: firmware updates, SDK dependencies, and how to balance rapid quality improvements with customer stability in an infrastructure-as-a-service environment.

Finally, we connect the dots to NAB season themes like global hardware supply chain pressure and why “getting off the cloud” often really means finding a more sustainable cloud pricing model.

Key topics

• One-year progress on Akamai Cloud VPU adoption and new VPU-backed VM SKUs
• Customer profiles including hyperscalers expanding beyond on-prem points of presence
• Live streaming and VOD workloads across premium and user-generated content
• Why Bitstreams matters as a GUI bridge for teams coming from managed services
• Partner ecosystem support for advanced requirements such as DRM
• The reality of firmware upgrade cycles in IaaS and how SDK dependencies affect rollouts
• New European footprint with London and Frankfurt for higher availability designs
• How to estimate VPU density using a 32× 1080p30 benchmark
• What NAB conversations reveal about supply chain pressure and cloud pricing sustainability

Stay tuned for more in-depth insights on video technology, trends, and practical applications. Subscribe to Voices of Video: Inside the Tech for exclusive, hands-on knowledge from the experts. For more resources, visit Voices of Video.

Transcript

Show Kickoff And Guest

Mark Donnigan, NETINT 0:07

Voices of video. Voices of video. The voices of video.

Voices of Video 0:12

Voices of video.

Mark Donnigan, NETINT 0:17

Welcome to this episode of Voices of Video. And today I am joined by Sarah Walter from Akamai. Sarah, it's great to see you again.

Sarah Walter, Akamai 0:28

Thank you for having me back, Mark. It's great to speak to you again.

Mark Donnigan, NETINT 0:31

I think this is our third. I should know this, but this is at least our third interview on Voices of Video.

Sarah Walter, Akamai 0:38

That's right.

Mark Donnigan, NETINT 0:38

I'm glad we're making it a I guess we did a webinar, so that kind of you know counts in there. Well, uh, you know, we were talking off camera before we started the recording, before we hit the record button, that it really was about one year ago that we began the the promotion to really talk about as a lead up to NAB what Akamai is doing with NetInts. And

One-Year VPU Check-In

Mark Donnigan, NETINT 1:03

I think a great place to start is kind of do a one-year check-in. Like what has happened from the Akamai perspective with the VPU accelerated instances on Akamai Cloud? And uh, you know, how are things going?

Sarah Walter, Akamai 1:17

Absolutely. I'm excited to catch up.

Mark Donnigan, NETINT 1:20

Yes.

Sarah Walter, Akamai 1:20

A lot is happening, as you know, um, a lot of growth and adoption among existing and new customers for both of us. And we're also about to double the number of VPU-backed VM SKUs that we have in market. We're introducing three new flavors of eight card plans to better suit our customers' needs. Um, and we've also just introduced VPUs into our London expansion data center. So a lot of exciting growth happening for the VPU product line here at Akamai.

Mark Donnigan, NETINT 1:55

Yeah. Yeah, you know, thank you for those updates. It's it is, it's super exciting. The the challenge, and this is what we talked about when we first made the announcement of Akamai integrating hardware, you know, into the Akamai cloud. Um the the challenge has always been that the industry, for the most part, is not hosting their own hardware, you know. So they need it sitting on a network somewhere, sitting in a cloud, they need it available. And uh for some, you know, that's just to manage, uh, you know, maybe they have kind of bursty demands. You know, they they they they don't want to have to make the CapEx investment. Others, it's just how they built their systems. They're like, we don't really operate data centers, you know, so we need it somewhere. I know that uh initially, you know, there was a lot of testing, there was a lot of evaluation, and you know, we were working very close together to get some of these deals, you know, kind of into production. But now I I think it's safe to say that it's off to the races, right?

Sarah Walter, Akamai 2:59

Absolutely. And we're seeing engagement coming from all sorts of places, obviously our mutual customer conversations and partnerships, but even some of these events and um industry events and things that we've been attending jointly, we've seen a lot of engagement there as well.

Mark Donnigan, NETINT 3:17

That's right. That's right. Well,

Customer Profiles And Use Cases

Mark Donnigan, NETINT 3:19

you know, I think it would be interesting. Can you give a profile of the types of customers or the use cases that are migrating to VPUs on Akamai?

Sarah Walter, Akamai 3:31

Yeah, I can give you maybe one interesting example of a deal we just signed with a and actually with a hyperscaler who's working internationally, who's already engaging and using VPU on-prem, but was looking for additional points of presence globally. And they were interested in the fact that we have VPUs available in Europe, in Frankfurt, now in London. And so that's been one value that we've been able to add for those customers is just offering additional locations and enabling them to expand beyond their on-prem offering.

Mark Donnigan, NETINT 4:08

That's right. Yeah, um this is a big deal. And it was really exciting for us because again, it was by having the hardware on Akamai that we could even service this customer. This particular hyperscaler, you know, has many VPUs installed in their network in their core operating regions or region. But as they were looking to expand, they obviously were not going to build data centers around the world, like you say, you know, in Europe or, you know, in North America, you know, Latin America, and various places. So, you know, without being in Akamai, on Akamai, unfortunately, they wouldn't have been able to continue to use VPU outside of their market. And so this was an incredible enabler. And are you seeing um, you know, interest? Are these live streaming workloads? Are is it VOD? Is it premium content? Is it sort of user-generated social? Like, are there any themes um, you know, across the customer base?

Sarah Walter, Akamai 5:11

I feel like we are seeing adoption from pretty much all of those categories at this point, which is it was a little bit of a leading question because obviously I know the answer, but as you know, I mean, we've had a number of these like larger customer engagements now in the last year, and they really do seem to span a variety of use cases. I think some of the themes we are seeing are customers and transition. So, again, this example is hyperscaler, looking for a global expansion, or we have customers, for example, who are moving away from more managed services, like Media Live and looking for something that is in between, something that's in between a true, fully managed stack that I'm paying for accordingly, versus hosting everything on-prem and managing my own infrastructure individually.

Mark Donnigan, NETINT 6:03

Yeah. Yeah. Well, um, I I think that's also a very notable development over the last year. When we first deployed, it was very much infrastructure as a service, was um Akamai provides at an incredible uh low hourly rate, starting at just 42 cents an hour for a VM with a single VPU. And I mean, which is just you know incredible how low that cost basis is. You know, I think um the the stream to run like a 24-7 and AP30, this could be AB1, H E B C or H264, it works out to just a little over $15 a month, which is just unheard of. There's absolutely nobody who can match that. But the challenge was in that first was it was you had to then bring your own software. You know, you had to, you had to either already have an application um or you had to go build it.

Bitstreams Makes VPUs Easier

Mark Donnigan, NETINT 7:04

What has changed over the last year is we ported bitstreams. And Sarah, you drove that. I can remember many calls, especially early on, saying everyone wants this, but it's a little bit like it's an engine, but they need to buy the car, you know, they need a car, you know. So can we can we solve this? And so how has Bitstreams been an enabler? I'm just curious from your perspective.

Sarah Walter, Akamai 7:26

Yeah, so for anyone listening who may not be familiar, Bitstreams is NetInt's own kind of single plane of glass control panel to operate your NetInt VPUs. And as you mentioned, I think that initially may have been a barrier to adoption for some customers, especially those coming from outside the VPU ecosystem, whether it's because you're right, they're coming off some sort of a managed type platform.

Mark Donnigan, NETINT 7:54

It's like a solution, you know, it's not like all the way to an OVP, but it's a solution. And right.

Sarah Walter, Akamai 8:01

And I mean, think about most traditional broadcast customers aren't used to interfacing with FFMPEG on a daily basis and using that to build out their stack, right? So we saw that barrier and we said, how can we make it easiest or easier at least for customers to use these cards on the Akamai platform? So in partnership with NetInt, what we've done is made that Bitstreams product available for free to Akamai customers to use. Meaning if you're deploying NetInt VPUs on the platform, you can also use that Bitstreams control panel to operate it and set up your transcoding stack. So it provides you with a graphical user interface, which is a really nice bridge, again, for folks who are coming from something like a managed service, more traditional broadcast, or just even outside of the VPU ecosystem, whether that's GPU or x86, it really lowers the barrier to entry and has been a great addition for a lot of customers.

Mark Donnigan, NETINT 9:07

Yeah.

Sarah Walter, Akamai 9:08

And I appreciate the partnership in making that happen.

Mark Donnigan, NETINT 9:11

Well, you know, thank uh thank you for pushing us because I mean, this was a product that existed. It's been in very active development for, well, two years now, I guess. But, you know, with customers like Akamai, and then obviously your end customer, it really helps us ensure that we're bringing the right solution. And also, I'd be remiss to not point out that you have also a set of partners, ISVs that are in the marketplace that have also done the integration work. So Bitstreams is is quite capable, very easy to use. People love the user friendliness, but you know, like it doesn't have DRM support, for example. Well, for some people, that's you know, that's absolutely required. Um, and and there's uh some other specific functions that you know certain sectors of the market would require, and it doesn't support that. But you have, you know, I'm thinking uh, you know, uh Capella systems. I know there was a a press release about the integration with Akamai and then also the support with NetEnt. So there are, you know, more fully developed solutions for those customers that need it, which I think is essential because, you know, especially with a new technology and with hardware, hardware is hard because, you know, it it's it's it's inherently fixed function, it's deterministic, and yet the software that goes around it requires somebody to have done the integration, to have understand, understood the wide variety of workloads, you know, that need to be supported. And the fact is, is that now Akamai has a really good ecosystem with the hardware engine of the VPU and the software. That's right. That's right.

Sarah Walter, Akamai 10:59

To your point, a full ecosystem of partners up and down the stack for transcoding and media in general. Um, and I've always appreciated NetInts partnership and attitude there, which has been very open and not just willing, but interested and actively pursuing working together with Akamai, with all of our partner ecosystem to serve customers however we we can best.

Mark Donnigan, NETINT 11:24

Yeah, that's right. Well,

Firmware Updates Without Disruption

Mark Donnigan, NETINT 11:26

and you know, I'm thinking too, maybe you can comment on this, but one of the one of the challenges that we had to get through was even like the firmware upgrade cycle. And that is always a challenge for an infrastructure provider. Now, you know, if you're only operating your own service, you know, then you're free to, you know, you can have your policies and you can have, you know, your methodology as to how you roll it out. But when you have customers, and you know, I'm thinking as just one example, even this mutual hyperscaler, because of the the work that we're doing with them in their core region, they have some specific requirements that that then Akamai had to ensure were available on their instances. So I know we cause you a little bit of consternation there, Sarah. I can remember you saying, these people are asking for updates and you know, like this is really frequent. So how did you I guess, you know, maybe I I think it actually would be useful for the listeners who might, if they're familiar with NetInt and they're thinking, yeah, but you know, is Ahmed going to be able to keep pace with the frequency? Can you comment on some of the things that you've done? Because I know you did have to kind of make some changes to your to to your firmware upgrade policies, et cetera. At least for us. I I I'm not commenting about the rest of your stack.

Sarah Walter, Akamai 12:52

That's true. It did take some some sorting out because, as you know, there are implications to upgrading the firmware for these devices for customers. In an IaaS environment and an infrastructure as a service environment, like most providers, we manage the firmware upgrades for our customers, meaning that the customers using the platform are also then beholded to whatever firmware version we have deployed. So we try and manage that, it manage that process in light of that fact and kind of in coordination both with you all as the vendor and producer of the firmware and our customers to make sure that we are actively addressing any security updates that come through, but also closely monitoring for these more customer-facing changes or improvements that are so important. The one thing we do keep in mind also is that there are SDK support implications for our customers. So, again, it's something we're trying to manage carefully. Um, and that means that the firmware version we deploy sometimes takes longer to deploy the kind of latest version that NetInt has released. But again, that is mostly done in coordination with kind of this customer feedback and making sure we're balancing between reviewing these new releases coming out and uh avoiding customer disruptions as we manage that process. So it has been a learning process just because that whole, you know, the SDK piece and all of that has been new for our customers versus other managed products like our GPU offerings. But that has been, I think, something we've we've gotten now to a really good place where we're managing to check all of those boxes across the board.

Mark Donnigan, NETINT 14:41

Yeah. Well, we really appreciate Akamai's flexibility because that that was one of the the the initial learnings, or maybe even a little bit of a speed bump that we had to get over. I I I think I can even remember the call where you said, we need to talk about the frequency of firmware updates. And, you know, and it was a learning because what is unique about NetEnt VPUs, and now the industry knows this, but traditionally, when silicon comes out of the fab, you know, comes out of the ground, does what it did at that time, and it really never changes. You know, occasionally there's some security patch or there's something, but basically, the especially on the encoder side, the rate control, the way that the structure of the encoder operates, you can't change it. What's very special, and it's what the industry really likes about the VPU, is that you can improve quality over time. So, like our HEVC efficiency quality has improved nearly 20% from when we first shipped the product. Added, you know, CRF rate control after the fact. And there's just, I mean, those are just two things, but you know, no one would ever think that's even possible to do that. So these firmware updates that are coming, these aren't just sort of minor, you know, bug fixes, security patches, that not the security patch is minor, but you know, these, you know, this really has impact. So these customers are like, no, we have to be running this version because that's you know, that's what we require. And uh so it's been wonderful. You know, thank you for working with us and developing this process because uh, you know, it really is essential.

Sarah Walter, Akamai 16:29

Of course, and I I know I speak for everyone on the Akamai side when I say we appreciate Netent's bias towards improving the customer experience continually, because that's something we try and do too. So that's information for sure.

Mark Donnigan, NETINT 16:43

Actually, we have internal uh OKRs around that. Uh the engineering team does, you know, it's lit, which you know, how unusual. Um, Quadra first started shipping uh at mass production, you know, four years ago or so. So imagine four years after shipping a chip, there still is no KR benchmark of a certain level of quality improvement. Like that's unheard of. You know, the the assumption is, oh, well, we would never want to adopt silicon because it's fixed function and it's a four-year-old encoder. How good can it be? Everybody else has gotten better. It's like, well, you know. So yeah, it's amazing. Well, Sarah, you teased out that you have a new data center going live. And

London Launch And Eight-Card SKUs

Mark Donnigan, NETINT 17:32

why don't you first of all give an overview of where VPUs are installed just regionally? And then you also teased out that there's some maybe new plans that are coming. So to the extent that uh, in fact, probably by the time this even goes live, those plans will have been announced. But to the extent that you can talk about it, you know, talk about the coverage expansion and then talk about these plans, if you would.

Sarah Walter, Akamai 18:00

I will. So the coverage expansion is exciting. As I mentioned, we have recently launched a new point of presence in our London expansion data center. So that's one for Slugwise. And the goal here was we had a number of European customers who were looking to create kind of a high availability setup with their VPU deployments in Europe. Um, and so they were we had a number of customers who were looking for a second point of presence in addition to Frankfurt. So now we are able to offer both London and Frankfurt in a MEA, which is extremely exciting. And that means we also have we have the two points of presence in the United States and now in Europe as well for customers looking to build that kind of configuration. As you mentioned, I also teased that we have new plans coming. So in addition to the three plans that we offer today, which we have two varieties of plans that have one card each, different sizes, and a two-card plan. We are set to introduce, hopefully, by the time this airs, three new plan sizes, each with eight cards, a small, medium, and large variety. And that is also just born out of customer feedback. We had a number of customers who were looking to do a lot on one VM. And for that, they wanted a lot of power. They wanted a lot of VPUs, they were looking for a lot of parallel streams, a lot of compute. And so this was our response to that was to create these new plans, and that will bring our total number to six.

Mark Donnigan, NETINT 19:40

Amazing.

Capacity Math And Cost Per Stream

Mark Donnigan, NETINT 19:41

And for the listeners, I'll I'll just I'll just give a really quick overview of what exactly that means when we talk about a one-card in terms of density, a one card, a two-card, and eight card solution. So each VPU, each quadra VPU can process, if you want to do the math, it's two billion pixels. So if you, you know, if you just want to run some easy calculations and whatever your core, your resolution or your resolution ladders are, just add up how many pixels um, you know, that would consume, and then you know, divide by two billion, and that would give you like the number stream. So if you're talking like an adaptive bitrate ladder, you know, you just you know add up the number of pixels that are in that adapt that all the resolutions and divide by two, you know, or divide that into two billion, and that would give you. But the easy benchmark that we always quote just because it's it's a reference, is a single VPU can encode 32 1080p 30, 30, so 30 frames in either AV1, H E B C or H264. It's the beautiful thing about silicon, is unlike software, where you have to say, well, wait a second, if it's you know uh AV1, it's a much, much, much more complex encoder. So your throughput is gonna go way down. Silicon, it doesn't matter. You don't get any any higher capacity with H264 or AV1. It's 32. So that means a single instance, um, the VPU would do 32, 1080p 30 channels. If you say, well, what about 60 frames? You just divide by two, so then it'd be 16. Um, 16, 1080p 60. So if you're doing live sports or something, um, two cards, you just double it, so that'd be 64, and then eight cards, you just do the math, and that'd be 256. Uh, so I mean, that is tremendous density. And that's why when you see some of the numbers that we're putting out there, you know, NetInt, we're you know, often talking about cost per stream, and we're putting a number of around $15 a month. That's how we arrive at that. You know, we're just basically looking at the cost of the VM, you know, and just doing the simple math. To uh to arrive at it. So hopefully that's helpful.

Sarah Walter, Akamai 22:04

That's a really helpful reference. And that was exactly some of the feedback that we've been getting from customers, is they're looking for that kind of density. They're looking for Yeah, in general at scale. So this is our response to that. Yeah.

Mark Donnigan, NETINT 22:19

And the beautiful thing too is, you know, with the small, medium, large, because the host machine, so the VPU, because the interface is NVMe, it's actually extremely low overhead on the host CPU. So if you're if you're really not using the host to do much more than just I/O, you know, transfer and network access, really just a very minimal processor, you know, x86 will work, will work just fine. But increasingly, you know, there are functions like de-interlacing that we don't support de-interlacing. Well, turns out that you know, you can do that on the CPU. And, you know, that could be one example. There might be, you know, other specific image processing. Maybe there's some subtitling functions. There can be can be certainly, or even just running the transcoding framework, the media processing framework on that. So with the small, medium, large CPU, and then with eight cards, we're pretty certain that the economics can't even be touched within a mile of any other solution.

Sarah Walter, Akamai 23:27

I happen to agree.

Mark Donnigan, NETINT 23:48

Amazing.

Sarah Walter, Akamai 23:49

So more details can be found on the website, obviously, but um excited about that introduction and what that means for our customers.

Mark Donnigan, NETINT 23:57

Yeah, that's uh that's great. Well, we're excited to um, you know, you also are listening to your customers. We appreciate you know you acknowledging that you know almost everything that we do at NetInt is customer driven, meaning that um, you know, we're just always out in the market saying, what do you need? You you know, what keeps you up at night? You know, what problems can we solve? And then, you know, we try and bring solutions to that. And I know even with this plan, because you know, from day one, we had even expressed to you, hey, you know, I think customers are gonna want beyond two cards. And, you know, it's like, well, let's get, let's, let's get deployments, let and then let's get feedback. And so it was very measured um in a good way, you know, the the response. So we're excited about that. Um,

NAB Trends Supply Chain And Cloud Costs

Mark Donnigan, NETINT 24:50

Sarah, we have NAB kind of around the corner. And, you know, I'm I'm curious as you are thinking and as Akamai is thinking about the um um the incredible changes, you know, the shifts in the industry, the shifts in the market, the, you know, there's some technologies are being actively embraced and other technologies are kind of falling by the wayside. And, you know, I'm just curious if you have any insights as to, you know, what your customers are really caring about today. You know, what kinds of things is, you know, are they going to be going to NAB, or if they're not at NAB, they're gonna be online, you know, looking for solutions, they're trying to solve the problem. And I'll let you answer that from a technical perspective, operational business, or all of it.

Sarah Walter, Akamai 25:44

I think the number one thing I'm hearing in customer conversations kind of across the board are concerns over global hardware, uh, this global hardware supply chain and what that means for product availability. Um, that has meant a lot of customers who've traditionally kept some workloads on-prem are now finding that expanding that is not really an option. Um, that they are looking to enter the cloud for reasons that they have not necessarily explored traditionally. Um, and it has become more of a necessity for a lot of our customers and they're looking to future proof in that sense. And so I think that is going to be maybe a driver in some of these conversations. Again, some of the customers who've more traditionally handled this kind of work and scale on premises are are going to be looking to transition to the cloud.

Mark Donnigan, NETINT 26:43

Yeah. You know, so I I think this is a really, really good point. And we've even, as NetInt, have had to sort of be really intentional in how we talk about the cloud, because there is an element of there are some mostly public clouds where the pricing is just getting unsustainable. And so you will sometimes in certain circles and certain sales conversations we're in, you know, hear even the direct words, we're trying to get off the cloud. But and and so if you just stop there, you'd be like, oh, whoa. So the industry is kind of maybe ping-ponging back to, you know, own data centers and things. And and some people are, and everybody has different strategies, right? But then when you just begin to probe a little bit more, you find out that no, no, they really don't have a desire again to go build a data center. In fact, they can't, you know, because of the you know, the capex involved, et cetera. What they're really saying is the unsustainable price models and business models of some um, you know, the public clouds, we have to find a s a response to that. Is that what you're seeing as well? Because it strikes me that Akamai, here here we are doing a whole episode and talking about how great the Akamai Cloud is. And yet I know that some of the joint customers that we have brought to Akamai initially, it was we have to get, again, in air quotes off the cloud. But really, what they're saying is we need to get something that's sustainable economically, you know, and they found that in Akamai. But I'm curious if if that level of distinction is also what you're running into.

Sarah Walter, Akamai 28:28

I have seen the same thing. And in a number of those cases, I think those customers have approached us looking to lock in commitments for longer-term engagements to navigate around that and ensure that the competitive pricing that we offer today is something that they can accumulate.

Mark Donnigan, NETINT 28:48

Doesn't change.

Sarah Walter, Akamai 28:50

Right. Because you know, these things, you know, they're always technically subject to change. But we are willing to explore those kind of commercial long-term right.

Mark Donnigan, NETINT 29:01

Long-term commitment is always going to be uh, you know, that's how all of our business works.

Sarah Walter, Akamai 29:07

We want to protect their continuity too.

Mark Donnigan, NETINT 29:09

Yeah, yeah, exactly. Yeah, and um, and you know, and look, and and you know, I'm not here to, you know, unpack this any further in terms of uh trying to make any particular other cloud vendor the enemy, but yeah, you know, everybody has their strategy, right? And certainly there's great lock-in with some of these, you know, some of these large, large platforms, but we are just seeing so universally, and it's again why the Akamai solution is just, we believe, perfect for the market because it provides, you know, yes, there's elements of it where it's not as complete from maybe, you know, if you want to say the microservices, you know, level. So, yes, it's definitely more infrastructure as a service, which means that the customer needs to be able or willing to in some cases do a little bit of work. But that gap is closing as to the work required. And yet the cost advantage is so overwhelming that we're just seeing just people who are saying, okay, you know, yes, we're getting almost everything from this other solution. And, you know, we just have to bolt a few things together, but otherwise we don't really have to do much. Over here, we have to do a little engineering, but the cost, the performance advantage is so overwhelming that people are moving in that direction. The industry is moving kind of away from these more monolithic, you know, platforms. So that those are some trends. So I just had to make that comment because you pointed out that you know, people are interested in moving to the cloud, and yet it's just fascinating. You know, the cocktail party talk at NAB is on the surface, high level, is going to be we've got to find a solution to the cloud. We've got to, and you're like, wait a second, you know, how can this, how can they both be true? It's like, well, so well, Sarah, I know that we could find lots of other very interesting things to talk about, but thank you again for coming on and giving us this one year, you know, one year update. I can't wait to make it a two-year anniversary and then we'll do a three-year. And so we're really happy to be working with Akamai. Thank you for your support. We really appreciate that.

Sarah Walter, Akamai 31:26

Thank you again for having me and for your support. That goes to the whole Neddit team and um our customers and partners who've also participated in this for the last year. I think it's been again, we've seen great growth both on the customer side, on the platform side, in terms of our engagement together. And I look forward to seeing what where we go next and what things we can cover in the next update.

Mark Donnigan, NETINT 31:52

Thank

Closing Invite And Sponsor

Mark Donnigan, NETINT 31:53

thank you for joining us. And if you want to come on the show, if you're doing anything interesting in video, feel free to reach out. We are very easy to find. I'm easy to find. You're watching this on LinkedIn or YouTube or on our website. Um, just reach out, tell me what you're doing. It does not have to be VPU specific. We are it needs to be in video, but you know, you're building something, you have some technology that you would like to tell the world about, tell me about it and let's see. Maybe we'll have you on and uh it'll be a great format for you. So thank you again. And until next time, happy encoding.

Voices of Video 32:36

This episode of Voices of Video is brought to you by NetInt Technologies. If you are looking for cutting edge video encoding solutions, check out NetInt's products at netint.com.