In this CDN Alliance-led session, Mark De Jong and Tomas Bacik explore the present and future of edge computing and its growing role in content delivery networks. Bacik outlines the most active edge use cases today—from real-time header manipulation and DRM enforcement to analytics and content personalization. While compute-heavy functions like storage and transcoding remain centralized, more responsive tasks are shifting to the edge to enhance performance and user experience.
Despite the promise, both experts acknowledge key challenges that hinder scalability—especially in multi-CDN environments. The biggest issue? Lack of standardization. “It adds complexity at many levels,” Bacik says. Without clear standards, even the most robust edge deployments face limitations in cost-efficiency and interoperability. Still, both agree that edge computing will become a core offering across CDN vendors. As WASM language support grows and more power is pushed to the edge, the industry must focus on collaboration and unified frameworks.
Mark De Jong:
Hi, this is MarK De Jong CDN Alliance. Welcome at the StreamTV Show. I'm sitting here with Tom. We just had a session about S-computing. I just want to do a couple of key takeaways that we had with the session that we had. So the first one is about use cases. So there are a lot of different use cases that you can do with S-computing right now. Not all of them are commercially viable right now to actually be used, but there are already a number of different use cases there is actually used and you have to have a lot of experience in this. Can you highlight a few of them?
Tomas Bacik:
I think basically I should start by saying that anything that needs to be done via config can now be done via code, which expands the pool of features we can support. But the centralized cloud and delivery model hasn't gone anywhere. It's still there and it's just that some type of operation are moving closer to end user. Usually those closely tied to delivery. So we see a lot of header manipulation, response adjustments, manifest customization being the most often one, probably related to video. Then there's a lot of using content protection, authorization, authentication, or even DRM. Then there's a big space for growth in content personalization. So we actually work with real-time data about user behavior, what devices they use, where they are to adjust content instantly based on that information. Apart from those features connected to content and request manipulation, we also see analytics, look, aggregation. So it might give us more visibility and intelligence into what actually going on at the edge. So that's the most common use cases we're seeing now, but the heavy liftings such as transcoding, encoding, storage, still stays in the cloud.
Mark De Jong:
Okay. Well, that is quite some couple of different use cases already that you can actually use. Another thing we touched on during the session was about challenges. I mean, multi-edge or edge in general is interesting. There are lots of capabilities with it. It can enhance a lot of stuff, but also has a lot of challenges, whether it's in development, the resources that you need to have, the knowledge, operations, management, the fact that you need to balance with the different CDNs and all having a different type of edge compute that they do. What do you see there in terms of challenges that you think we should work on?
Tomas Bacik:
I'd say firstly it adds complexity at many levels. So there's a lot of engineering effort that needs to go into that in both deployment and the management of edge computing. So it all comes down to TCO. You need to make sure the question always is whether the performance improvements actually justify the resources you put in. The lack of standardization would be the biggest drawback I see right now.
Mark De Jong:
Fully agree.
Tomas Bacik:
And especially in multi CDN scenarios, which are relevant to all large clients now. So I guess that slows the adoption.
Mark De Jong:
Okay. Okay, cool. Another thing we touched at the end of the session was about the future of edge compute. I think we all know it's going to stay, we just don't know how it's going to evolve over time. So if you have to paint a picture, look at your crystal ball, what would you say? Okay. That's the future of edge compute.
Tomas Bacik:
I'm sure edge computing is going to be the integral part of CDN's offerings, so it's going to stay there. That's been more power at the edge and we'll see more deployments. I think as an industry, we need to work harder on the standardization to support the adoption of multi-edge setups basically. But right now we already see wasn't based language support as a standard requirement by some clients. So I'm confident that we'll see more and the trend of CDN vendors actually becoming a full-fledged complex service organization will only continue with that.
Mark De Jong:
Okay, cool. Well, thank you Tom.
Tomas Bacik:
Thank you for having me.
Mark De Jong:
Good.