An Epidemic P2P Computing System and Its Application in High-Resolution Video Processing
Guowei Chen, Yanru Chen, Koji Kataoka, Md. Humayun Kabir, Takuro Sato · ITC-CSCC :International Technical Conference on Circuits Systems, Computers and Communications · 2009
Research on developing an epidemic P2P computing system for data-incentive applications was made, motivated by requirement of processing of high-resolution videos. The system employs techniques of epidemic data dissemination, status broadcast, and un-centralized task distribution, which brings features of efficient file transfer, automatic load-balancing, and non-existence of central failure points. The simulation results show remarkable improvement - with 51 workers, the proposed system costs 120 minutes in file transfer time and 224 minutes in task completion time, while a worker-centric scheduling grid system for data-incentive applications [1] costs 1020 minutes in file transfer time and 1222 minutes in task completion time.