| Grid computing systems, peer-to-peer services, and wide-area testbeds such as PlanetLab provide opportunities to collect a significant amount of distributed resources from various autonomous sites. However, effective resource federation for these systems remains a challenge; which in turn limits the further development of these systems. Traditionally, resource allocation policy has both been specified and enforced centrally. Unfortunately, there is an inherent mismatch when attempting to map such traditional techniques to inherently decentralized infrastructures. At the same time, federated infrastructures are more likely to consist of a number of self-interested users who, while not malicious, are certainly interested in maximizing their own allocation of global system resources. Thus, a scalable, decentralized, distributed resource federation system must be built to efficiently allocate resources for participants and to enforce fair resource allocation based on local policy with no central infrastructure or policy.; This thesis presents SHARP, a framework for secure distributed resource management in a global-scale computing infrastructure. The cornerstone of SHARP is a construct to represent cryptographically protected resource claims---promises or rights to control resources for designated time intervals---together with secure mechanisms to subdivide and delegate claims across a network of resource managers. These mechanisms enable flexible resource peering: sites may trade their resources with peering partners or contribute them to a federation according to local policies. In this Way, SHARP enables distributed and mutually distrustful entities to exchange computational resources in the same manner that ISPs exchange network bandwidth. A separation of claims into tickets and leases allows coordinated resource management across the system while preserving site autonomy and local control over resources. SHARP also introduces mechanisms for controlled, accountable oversubscription of resource claims as a fundamental tool for dependable, efficient resource management.; This thesis discusses a variety of resource peering policies that can be supported by SHARP. Further, the thesis presents complete solutions including protocols, algorithms; and implementations to support various peering policies on SHARP. We present experimental results from a SHARP implementation for PlanetLab, and illustrate the use of pair-wise resource peering as a basis for a decentralized bartering-based economy for global computational resources. The results demonstrate the power and practicality of the architecture. |
