Scalable Computing Infrastructure (SCOPE) Project

Discovery of Computing Resources: To support massively parallel computations mechanisms are needed that will allow the job manager to discover appropriate resources. This is the "resource discovery" problem applied to processing resources and the same techniques used for information discovery can be applied. System managers can register with directory servers, providing aggregate information about the resources they manage and the characteristics of the service itself such as reliability guarantees and pricing. Because this information changes, the use of a directory service that supports remote update is needed.

• Distributed authorization and payment: The client organization will maintain its own authorization database defining a group whose members can access the computing resources of the contracted service providers. The service providers lists the group in the authorization database and users would present authorization credentials proving membership in the group when they need to use computing resources. A distributed accounting server could issue credentials certifying that the client can pay for the service, and can provide payment to the service provider when the service is complete.
• Privacy and integrity: Many parallel computations require access to and manipulation of sensitive information. Cryptographic techniques and checksums may be used to protect against such threats.
• Assurance for selection of compute services: Data is also vulnerable on the processing node itself. It possible for the user to specify assurance criteria for selecting compute services.