Technology review: evaluating peer-to-peer solutions for online courses and programs

Community College Enterprise, The, Spring 2008 by Klinker, Eric

As community colleges plan to incorporate rich media into their courses and programs, the costs and benefits of traditional, centralized content delivery networks (CONs) versus peer-to-peer (P2P) networks emerge as a key issue. Here's a look at which applications work best with P2P, how to leverage P2P technology alongside traditional means of delivery, like CDNs, and how to evaluate these technologies for any number of current and emerging applications.

The demand for rich media delivered via the Internet is exploding in the realm of education as well as entertainment and business as the quality of content online improves dramatically and reaches ever wider audiences. To meet this growing demand, many new solutions for content delivery are coming to market, with a growing number of innovative new solutions, such as P2P and hybrid content delivery networks (CDNs*) being offered alongside traditional CDNs. While community colleges, have for the most part, stuck with traditional CDNs for delivering their content, P2P has the potential to upend the economics associated with enabling high-quality video delivery. P2P content delivery holds such promise that if you don't already have a P2P strategy, rest assured your competition does.

Peer-to-peer overview

As a starting point, it is helpful to understand how peer-based models differ from the traditional content delivery models. In the traditional CDN model, the intelligence and capacity of the system is centralized and located in a core set of servers, while clients are simply mute consumers in the delivery process. The opposite is true in peer-based systems, where users contribute content they have already consumed back to other users of the application. This allows considerable intelligence to be located at the edge of the network, within the peers, and distributes the resources required for content delivery throughout the internet. The result is a far more efficient delivery model at scale, but it does not guarantee a minimum level of resource availability, as is the case in the centralized CDN. Conversely, the CDN cannot guarantee enough infrastructure is deployed at any given time or place to meet high levels of demand. In the peer model, additional infrastructure is "created" on the spot with high demand in the form of more users. Such are the pros and cons associated with each model.

Using a film clip in a single section of a history or nursing class for example, is simply a matter of having technology appropriate to the size of the room and number of students: a VCR and large television monitor, a computer with a DVD monitor and ceiling mounted projector.

Having seven sections of History 138 (Contemporary World History) view Judgment at Nuremberg (1961) along with Nuremberg Triais (1947 - a Soviet documentary) presents much more complex problems. Similarly, having multiple sections of Nursing 213 (Adult Medical - Surgical Nursing 3) view film clips of a surgical procedure and review pre- and post-op x-rays requires a system robust enough to support multiple simultaneous accesses. While asynchronous modalities (such as online) provide some buffering, the increase in online offerings and enrollments now presents similar problems to synchronous modalities such as face-to-face or interactive video.

To harness the strengths of each, commercial solutions are now coming to market that combine elements of traditional CDNs with P2P technology to create a "hybrid" or "peer-assisted" CDN. While there are many categories of pure play P2P technologies, this review will focus on the hybrid offerings, beginning with the typical applications of content delivery.

When is P2P appropriate?

Peer-based systems work best when a large number of peers are available to exchange the content being accelerated. Of course, you need only one peer with content to make the system function, as is often the case with personal communication. After the first peer, there is a general threshold around peer availability (in terms of number and proximity of those sharing the content) that must be crossed before a system is most efficient. The efficiency threshold varies by technology; but generally speaking, the more popular a file is, the more likely available peers will have downloaded it and, in turn, the more effective the peer system is at delivering the content.

Similarly, because peer-based systems involve some overhead in starting up peer communication, they often don't make sense for very small file delivery. If something can be delivered in a few seconds over Hypertext Transfer Protocol (HTTP), it's probably too small for peer-based delivery. As a point of reference, if the P2P startup overhead is 5 seconds and the users are downloading at IMbps, then the object size threshold is under 1MB. Conversely, as the file size increases, the overhead (as a percentage of download time) is minimized and the more efficient a peer system becomes.

Because the effectiveness of peer-based systems will vary depending on the popularity and size of an object, they are typically deployed alongside dedicated servers or traditional CDN services to create a hybrid, or peerassisted CDN solution. It affords a degree of quality assurance for long tail (less than popular) objects or newly available objects and guarantees that each object can be delivered at the appropriate data rate.