Title :
A P2P approach for global computing
Author :
Dou, Wen ; Jia, Yan ; Wang, Huai Ming ; Song, Wen Qiang ; Zou, Peng
Author_Institution :
Dept. of Comput. Sci., Changsha Inst. of Technol., Hunan, China
Abstract :
We describe a peer-to-peer self-organizing overlay network for our global computing system. The preliminary simulation results show that the network has some small-world characteristics such as higher clustering coefficient and short path length, which leads to an efficient heuristic task scheduling algorithm on which any volunteer peer with limited knowledge about the global network can dispatch its excrescent computation tasks to powerful nodes globally, in a way contrary to the current global computing system in which a global broker is responsible for the task scheduling. We argue that our approach is a starting point to eliminate the broker component which makes current global systems unscalable.
Keywords :
Internet; grid computing; scheduling; self-adjusting systems; P2P approach; clustering coefficient; excrescent computation tasks; global computing system; heuristic task scheduling algorithm; limited knowledge; path length; peer-to-peer network; powerful nodes; self-organizing overlay network; Computational modeling; Computer architecture; Computer networks; Computer science; Load management; Military computing; Peer to peer computing; Processor scheduling; Scalability; Scheduling algorithm;
Conference_Titel :
Parallel and Distributed Processing Symposium, 2003. Proceedings. International
Print_ISBN :
0-7695-1926-1
DOI :
10.1109/IPDPS.2003.1213451