The State of the Stack
Live streaming places a different and more sophisticated load on the network designers and operators: sufficient capacity with flexibility in configuration, but at the same time not over provisioned/underutilised to the extent that it is uneconomical. Live streaming architecture is very involved with telecoms and not just limited to store and forward internet content delivery between servers and eventually end users. Tight Layer 3 (IP) network operations are therefore at the core of live streaming offerings at scale. CDNs still buy from Layer 2 and Layer 3 telcos, and the telcos buy into peering facilities. The more they buy, the lower their direct cost base.
Today, good, low-cost IP connections are abundantly available, and the trend is towards longer backhaul of IP from users to fewer, better-connected, centralised data centres managing the session and application and keeping the majority of the functionality in one place. In its pure form, this has been nicknamed the "death star" architecture, although it is more properly/ traditionally called a "hub and spoke" architecture. The downside is that a local problem at the hub takes out everything where in a distributed architecture there are fewer risks of centralised points of failure. The upside is you can centralise and optimise all your maintenance and support issues. It's worth noting that, in all high-availability CDN service architectures, the hub itself is usually a geographically distributed range of services clustered to act like a single central hub to ensure continuity in the event of even the most significant problems.
Telcos, ISPs, CDNs, OVPs, and Workflows
I find this trend towards recentralisation interesting, not least because CDNs were once put in place to reduce long-haul telco data charges (which were more expensive than shorter ones) by using proxies to cache or split streams nearer their audiences, lowering the overall cost. Now CDNs have added other layers of service, creeping up or down their natural location in the stack. These services are specific to the CDN, differentiating themselves more from the Layer 2/3 telco/ISP; this means that simply stating, "We have lots of servers everywhere," is no longer the key performance indicator it once was for buying decision makers (telcos and ISPs can make much the same boast). It's now all about those extra features that make them "more than an ISP" in that they offer CDN-specific services (which ISPs typically don't), and yet, at the same time, they tread a difficult line with their own customers, who tend to be about offering a host of professional services to do with content production workflows in addition to simple hosting of the origin or source content. This work is falling more and more to the online video platforms (OVPs) and the many boutique streaming media and webcast service providers. By creeping up the stack, the CDNs often compete with niche service providers that are using the CDN for lower-layer services.
Even more interesting is the number of telcos that are now internally deploying CDNs, which is ironic, given the view that CDNs were put in place to bypass telco pricing. It implies that telcos are finding it cheaper to circumnavigate their own pricing internally by deploying a CDN rather than simply managing their network better. Even more ironic is the number of telcos that are effectively buying their own bandwidth back again from the CDN; Akamai/BT and Global Crossing/Limelight come to mind here.
Companies and Suppliers Mentioned