In this study, we propose a processor allocation algorithm. Its concept is to lead the job scheduling from homogeneous systems into heterogeneous systems. In the computing process, the algorithm in this study will convert the computing ability of every PE (MIPS, the unit of the computing ability of a PE) into the amounts of job time units they provide every second, and gather the amounts of job time units of PEs in a resource. The gathered job time units of different resources can be treated as different independent computers. Therefore, adding the communication cost model into grid systems, the use of algorithm proposed in this study can get following results: obtain the job’s earliest finish time, reduce the unnecessary communication cost, shorten a list of jobs completion time, and increase the utility rate of systems. According to the experiment results collected from the experiment in the chapter 5, each resource provides computing power and the completion time of a list of jobs is close to the average finish time of 50 lists of jobs. Finally, according to the experiment results, we can find the schedule length is better than MEET, when the processor allocation algorithm is adopting the algorithm proposed in this study.
In the future, some extensive discussions from this study are listed as following:
¾ Job splitting: the objective of this study is focused on transforming job scheduling from homogeneous systems to heterogeneous systems, and therefore the job splitting is not considered. In the future, it is practicable to split a job into many independent sub-jobs, and dispatch these sub-jobs to suitable resources. In additional, it is considerable to adding load balance for jobs.
¾ Dynamic scheduling: modifying the static scheduling into dynamic scheduling because of the new job appearing in the system at any time. Dynamic scheduling can pay close attention to the new job immediately to give the dynamic scheduling service and broaden the scope of services.
¾ Multi- Grid-Schedulers: this study focuses on a kind of centralized scheduling process. Through a centralized Grid-scheduler the jobs submitted by the user will be gathered and allocated to the distributed system after the estimation. There may be
some bottlenecks of centralized scheduling at the Grid-Scheduler, but the distributed Grid-Scheduler will be a better method for the scheduling strategy when the loading, which needs to be scheduled, is quite heavy, or in the dynamic scheduling.
References
[1] Bailey, D., et al., “The NAS parallel benchmarks,” RNR-94-007, NASA 1994.
[2] Buyya, R., and Murshed, M., “GridSim: A Toolkit for the Modeling and Simulation of Distributed Resource Management and Scheduling for Grid Computing,” The Journal of Concurrency and Computation: Practice and Experience (CCPE), Volume 14, Issue 13-15, Wiley Press, Nov.-Dec., 2002.
[3] CFD Online, http://www.cfd-online.com/.
[4] Computational fluid dynamics,
http://en.wikipedia.org/wiki/Computational_fluid_dynamics.
[5] Ernemann, C., Hamscher, V., Schwiegelshohn, U., Yahyapour, R., and Streit, A, 2002, “On Advantages of Grid Computing for Parallel Job Scheduling,” In Proceedings of the 2nd IEEE/ACM international Symposium on Cluster Computing and the Grid (May 21 - 24, 2002). CCGRID. IEEE Computer Society, Washington, DC, pp. 39-46.
[6] Foster, I., “What is the Grid? A Three Point Checklist,” GRIDToday, July 20, 2002.
[7] Foster, I., and Kesselman, C., The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann, San Francisco, CA, 1999.
[8] Foster, I., Kesselman, C., and Tuecke, S., “The Anatomy of the Grid: Enabling Scalable Virtual Organizations,” The Intl. Journal of High Performance Computing Applications, 15(3): pp. 200-222, Fall 2001.
[9] Gehring, J., and Preiss, T., “Scheduling a Metacomputer with Uncooperative
Sub-schedulers,” In Proceedings of the Job Scheduling Strategies For Parallel Processing, Lecture Notes on Computer Science Vol. 1659, pp. 179–201, San Juan, Puerto Rico, April 1999.
[10] GridCafe, “全世界的網格計畫,”
http://www2.twgrid.org/gridcafe/projectworld/projects.htm.
[11] Guo Leitao, “Grid,” http://www.chinagrid.net/grid/paperppt/USTC/Ltguo.ppt.
[12] Hamscher, V., Schwiegelshohn, U., Streit, A., and Yahyapour, R., “Evaluation of Job-Scheduling Strategies for Grid Computing,” In Proceedings of the First IEEE/ACM international Workshop on Grid Computing, Lecture Notes on Computer Science Vol.
1971, pp. 191-202, Springer-Verlag, London, 2000.
[13] Howell F, and McNab, R., “SimJava: A discrete event simulation package for Java with applications in computer systems modelling,” Proceedings of the 1st International Conference on Web-based Modelling and Simulation, San Diego, CA. Society for Computer Simulation, 1998.
[14] Introduction to Computational Fluid Dynamics (ie CFD), http://www.cham.co.uk/website/new/cfdintro.htm.
[15] Introduction to quantum mechanics,
http://en.wikipedia.org/wiki/Introduction_to_quantum_mechanics.
[16] LCG Project Overview, http://lcg.web.cern.ch/LCG/overview.html.
[17] Li, K., “Job scheduling and processor allocation for grid computing on metacomputers,”
Journal of Parallel and Distributed Computing, Vol. 65, Issue. 11, pp. 1406-1418, November 2005.
[18] List scheduling - Wikipedia, the free encyclopedia, http://en.wikipedia.org/wiki/List_scheduling.
[19] Quantum computer, http://en.wikipedia.org/wiki/Quantum_computer.
[20] Quantum computing, http://eve.physics.ox.ac.uk/Personal/steane/qcomp.html.
[21] SETI@home, http://setiathome.berkeley.edu/.
[22] Silberschatz, A., Galvin, P. B., and Gagne, G., Operating System Principles, 7th ed, Hoboken, NJ : J. Wiley & Sons, Inc., 2006.
[23] SimJava, http://www.dcs.ed.ac.uk/home/hase/simjava/.
[24] Smarr, L., and Catlett, C.E., “Metacomputing,” Communications of the ACM, Vol. 35, Issue. 6, pp. 44-52, June 1992.
[25] Stibor, M., and Kutil, M., “Torsche Scheduling Toolbox: List Scheduling,” In
Proceedings of Process Control 2006, Pardubice: University of Pardubice, June 13-16, 2006.
[26] Sulistio, A., Cibej ,U., Robic ,B., and Buyya, R., “A Tool for Modelling and Simulation of Data Grids with Integration of Data Storage, Replication and Analysis,” Technical Report, GRIDS-TR-2005-13, Grid Computing and Distributed Systems Laboratory, University of Melbourne, Australia, Nov. 8, 2005.
[27] Sulistio, A., Poduval, G., Buyya, R., and Tham, C. K., “On Incorporating Differentiated Levels of Network Service into GridSim,” Future Generation Computer Systems (FGCS),
Amsterdam, The Netherlands, May 2007.
[28] Sulistio, A., Yeo, C. S., and Buyya, R., “A Taxonomy of Computer-based Simulations and its Mapping to Parallel and Distributed Systems Simulation Tools,” International Journal of Software: Practice and Experience, Volume 34, Issue 7, Pages: 653-673, Wiley Press, USA, June 2004.
[29] Taiwan UniGrid Project, http://www.unigrid.org.tw/index.html.
[30] Tanenbaum, A. S., and Woodhull, A. S., Operating Systems: Design and Implementation, 3rd ed, Upper Saddle River, N.J., Pearson Prentice Hall, 2006.
[31] The GRIDS Lab and the Gridbus Project, http://gridbus.csse.unimelb.edu.au/gridsim/.
[32] The Network Simulator - NS-2, http://nsnam.isi.edu/nsnam/index.php/Main_Page.
[33] The SimOS Home Page, http://simos.stanford.edu/.
[34] TIGER Grid Broker, http://gamma2.hpc.csie.thu.edu.tw/.
[35] TIGER Grid, http://gamma2.hpc.csie.thu.edu.tw/ganglia/.
[36] Wang, C.M., Chen, H.M., Chang, C.C., and Wu, J.J., “A High-Performance Virtual Storage System for Taiwan UniGrid,” Proceedings of the Third Workshop on Grid Technologies and Applications, CHU, Hsinchu, Taiwan December 7-8, 2006.
[37] Welcome to the SimGrid Project, http://simgrid.gforge.inria.fr/.
[38] World Wide Grid(WWG), http://www.gridbus.org/ecogrid/wwg/.
[39] Yahyapour, R., “Design and evaluation of job scheduling strategies for grid computing,”
Doctoral Thesis, Computer Engineering Institute, Lehrstuhl für Datenverarbeitungssysteme, November 2002.
[40] Yang, C.T., and Ho, H.C., “A Shareable e-Learning Platform Using Data Grid
Technology,” In Proceedings of the 2005 IEEE international Conference on E-Technology, E-Commerce and E-Service (Eee'05) on E-Technology, E-Commerce and E-Service, EEE, IEEE Computer Society, Washington, DC, pp. 592-595, March 29 - April 01, 2005.
[41] Yang, C.T., Chen, S.Y., Chen, T.T., Yeh, Y.C., and Chen, C.Y., “Design and
Implementation of Monitoring and Information Services Using Ganglia and NWS for Grid Resource Broker,” Proceedings of the Third Workshop on Grid Technologies and Applications, CHU, Hsinchu, Taiwan December 7-8, 2006.
[42] 林誠謙, 李世昌, 鄧炳坤, “下一代網際網路新紀元－全球網格(World Wide Grid),”
自然科學簡訊, 第十五卷第四期, pp. 123-124, November 2003.