A stock options metaphor for content delivery networks

Abstract

We focus on the timely issue of Content Delivery Network (CDN) resource management. We introduce a multi-stage scheme that leverages solutions from the capital market. The mechanism of Stock Options (SOs) is used to address a potential scarcity of resources, not adequately addressed by other predictive mechanisms previously introduced by the authors. Using a Predictive Reservation Scheme (PRS), network resources offered by a CDN are monitored through established techniques (Kernel Regression Estimators) in a given time frame. Next, a Secondary Market (SM) significantly reduces resource waste by allowing the fast exchange of unused (remaining) resources granted to Origin Servers (OSs). This exchange occurs either by implementing socially optimal practices or by allowing automatic electronic auctions at the end of the day (EoD) or shorter time intervals. Finally, we further enhance our Load Prediction Mechanism (LPM); SOs are purchased and exercised, depending on the lack of resources at the EoD. As a result, OSs may acquire resources (if required) at a standard price. The effectiveness of the proposed stock market-based CDN resource management framework further improves.