Skytide Blog

In the bleak mid-winter, it may be difficult to mentally grasp how soon that the 2012 Summer Olympics will be underway.  In just 164 days, nearly a million fans from every corner of the globe are expected to begin streaming into London for the festivities, pushing an already congested metropolis to the limits of its capacity.

A similar scenario will play out online.

Millions of fans will log on to their broadband service to view the action, creating bandwidth bottlenecks brought about by peak demands on the network.

Content Delivery Networks (CDNs) will have their hands full.

They’ll need to accommodate the huge traffic spikes or risk angering their customers.   The problem with provisioning bandwidth to this type of peak usage, however, is that CDNs are left with excess capacity during off-peak times.

Infrastructure-intensive industries that face similar fluctuations in capacity utilization — like electric utilities and public transit — are increasingly turning to peak load pricing. To reduce gridlock, the city of London even imposes a “congestion charge” on vehicles that travel through specific zones during peak times.

CDNs need to put something similar in place.  But doing so will require a different approach.

Most CDNs currently factor peak usage into their pricing using “95th percentile billing,” where they charge content providers based on an approximation of their peak bandwidth utilization.  That method is deeply flawed though, particularly because it does nothing to motivate a CDN’s customers to shift usage to the network’s off-peak times

We’ve just published a new whitepaper that addresses this issue in more depth and puts forward a solution for CDNs to adopt peak load pricing.

Read Improving CDN Capacity Utilization with Peak Load Pricing