Overview
The Peer-to-Peer Arhictecture for Data-Intensive Cycle Sharing (ADICS) is P2P data sharing software architecture currently being developed as part of the EU FP7 EDGeS project with previous support from EPSRC grant EP/C006291/1. ADICS is working to provide a simple means for scientic applications, especially volunteer computing projects such as BOINC that leverage idle CPU power, to also take advantage of the network and storage resources available on the network.
ADICS is made up of three main elements:
- a client application that allows for the downloading of data from peers on the network
- a data serving (i.e., caching) application that replicates data on the network
- a metadata lookup service that keeps track of which peers have individual data items
ADICS Network
That ADICS network differs from many other Peer-to-Peer networks in that it allows for the explicit separation of the data-sharing nodes from other participants in the network. This customizable provisioning of the network has a three-fold benefit to general applications and scientific volunteer computing projects in general:
- it allows for security criteria and replication strategies to be imposed upon data sharing peers
- it issolates the clients on the network from participating in the data-sharing overlay should they choose not to
- it provides a mechanism for central servers to offload data distribution to a trusted overlay
ADICS Features
- Provides a buffer between data producers and data consumers
- File swarming to maximize data propigation and client-download thoroughput
- HTTP interface to support legacy application integration and browser downloads
- Employment of X.509 credentials for authentication and authorization
- Support for grainular security policies to suit project requirements
- Fast data lookup mechanisms, with the DHT implementation at O(logN) steps
