Key Research Highlights > Cloud Computing > Peer-to-Peer-based Cloud Data Management
Kian Lee TAN |
Peer-to-Peer-based Cloud Data Management
Objective Integration of cloud computing, database and P2P technologies to exploit the advantages of P2P architecture on a cluster of nodes over MapReduce framework which include: more robust query processing mechanisms as nodes can communicate with one another; removal of single point-of-failure in the master/slave architecture of MapReduce; and facilitation of elastic design as peers can be readily added and removed. Results We have met the objectives of this project through the development of 1) BestPeer++, a data management system with distributed access control, multiple types of indexes, and pay-as-you-go query processing for delivering elastic data sharing services in the cloud, and instances organised as a structured P2P overlay network; and 2) the Katana framework, a novel P2P-based generalised data processing framework that can be deployed on many of the currently know structured P2P overlays that provides a programming model in which processing logic may be implicitly distributed with universality and expressiveness, and where kata and ana functions can be executed independently from each other. |