Scheduling Strategy of P2P Based High Performance Computing Platform Base on Session Time Prediction

Authors:
Hao Zhang;Hai Jin;Qin Zhang
Affiliations:
Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China 430074;Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China 430074;Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, China 430074
Venue:
GPC '09 Proceedings of the 4th International Conference on Advances in Grid and Pervasive Computing
Year:
2009

Citing 11
Cited 1

Chord: A scalable peer-to-peer lookup service for internet applications

Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
A scalable content-addressable network

Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications
SETI@home: an experiment in public-resource computing

Communications of the ACM
Skip Lists: A Probabilistic Alternative to Balanced Trees

WADS '89 Proceedings of the Workshop on Algorithms and Data Structures
Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems

Middleware '01 Proceedings of the IFIP/ACM International Conference on Distributed Systems Platforms Heidelberg
Improving Scheduling of Tasks in a Heterogeneous Environment

IEEE Transactions on Parallel and Distributed Systems
BOINC: A System for Public-Resource Computing and Storage

GRID '04 Proceedings of the 5th IEEE/ACM International Workshop on Grid Computing
SkipNet: a scalable overlay network with practical locality properties

USITS'03 Proceedings of the 4th conference on USENIX Symposium on Internet Technologies and Systems - Volume 4
WaveGrid: a scalable fast-turnaround heterogeneous peer-based desktop grid system

IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
Cluster computing on the fly: P2P scheduling of idle cycles in the internet

IPTPS'04 Proceedings of the Third international conference on Peer-to-Peer Systems
Tapestry: a resilient global-scale overlay for service deployment

IEEE Journal on Selected Areas in Communications

High throughput computing over peer-to-peer networks

Future Generation Computer Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

P2P based high performance computing (HPC) system introduces many new and interesting problems. P2P environment is heterogeneous and asynchronous. At the same time, P2P platform is not stable. The joining and leaving of peers are random. These characteristics make the P2P based HPC platform have great difference to the traditional HPC platform and the global computing project. To achieve effective job scheduling on P2P based platform, this paper introduces a DHT based monitor and task management scheme. Further, we propose a data structure of distributed bidirectional Skiplist to keep the prediction session time. Our scheme distributes the task to the nodes which have longer online session time. With such scheme, we can reduce the migration of tasks among different nodes and improve the resource utilization of computing nodes. Finally, we use a real trace to demonstrate the efficiency of our algorithms and scheduling schemes.