Scheduling Policies for Enterprise Wide Peer-to-Peer Computing

Peer-to-Peer (P2P) computing is a distributed computing paradigm which considers unreliable peer collections as target platform for application execution. In this paper such collections are considered in the context of enterprises. Unlike other P2P scenarios, more control and knowledge is available in regards to peers. This paper focuses on application task scheduling for such environments. Environment and application characteristics are identified and used to develop a set of characteristics for task scheduling policies. This characteristic set is then used to survey scheduling policies in similar distributed computing paradigms to find their relevance for enterprise P2P computing. A proof of concept simulation is presented for selected scheduling policies to outline some of the implementation decisions and validate the enterprise P2P computing paradigm.