An important factor that needs to be considered by every Grid application end-user and systems (such as schedulers or mediators), during Grid resource selection and mapping to applications, is the performance capacity of hardware resources attached to the Grid, and made available through its Virtual Organizations. In this paper, we represent the performance of a computational Grid as a regression model that can be used to fine-tune the selection of suitable Grid resources. A study on the performance of distributed systems with respect to particular variations in parameters is presented. Our objective is to use a measurement-based evaluation technique to characterize the specific performance contribution of the individual Grid resource configurations. In the process, we identify the key primary parameters (or factors) that should be considered when selecting and allocating a computational node for user application execution.
CollinsL., DziakJ. and LiR., Design of experiments with multiple independent variables: A resource management perspective on complete and reduced factorial designs, Psychological Methods14(3) (2009), 202.
2.
ChenV., BartonR. and AllenJ., COSMOS technical report: A review of design and modeling in computer experiments, Article published in Handbook of Statistics Elsevier Science 22 (2003), 231-261.
3.
TsouloupasG. and DikaiakosM.D., GridBench: A tool for benchmarking grids, Proceedings of the 4th International Workshop on Grid Computing (Grid2003), IEEE Computer Society Press: Los Alamitos, CA, (2003), 60-67.
4.
JaynesJ., Contributions in design of experiments: Methods and applications, Doctor of Philosophy in Statistics thesis University of California, Los Angeles, 2013.
5.
EzugwuA.E., FrincuM.E. and JunaiduS.B., Performance characterization of heterogeneous distributed commodity cluster resources, in: Adaptive Science & Technology (ICAST), 2014 IEEE 6th International Conference on IEEE, (2014), 1-8.
6.
CoulourisG., DollimoreJ., KinfbergT. and BlairG., Distributed systems: Design and concepts, fifth edition, Pearson Education, Inc., publishing as Addison-Wesley, 2012, p. 2.
7.
MyersR., MontgomeryD. and Anderson-CookC., Response surface methodology: Process and product optimization using designed experiments, 3rd edition, Wiley, New York, 2009.
8.
MontgomeryD., Design and analysis of experiments, 7th edition, Wiley, New York, 2009.
9.
MeeR., A comprehensive guide to factorial two-level experimentation, Springer, New York, 2009.
10.
WuC. and HamadaM., Experiments: Planning, analysis and parameter design optimization, 2nd edition, Wiley, New York, 2009.
11.
CaoJ., SpoonerP., JarvisA. and NuddR., Grid load balancing using intelligent agents, Future Generation Computer Systems21(1) (2005), 135-149.
12.
DouglasC., Design and analysis of experiments, John Wiley & Sons, 2008.
13.
Globus: MDS document: www.globus.org/mds [2014].
14.
WolskiR., Dynamically forecasting network performance using the network weather service, Cluster Computing1(1) (1998), 119-132.
15.
PengL., SeeS., SongJ., StoelwinderA. and NeoH.K., Benchmarks performance on cluster grid with NGB, Proceedings of the High-Performance Grid Computing Workshop, 18th IEEE International Parallel and Distributed Processing Symposium (IPDPS), Santa Fe, NM, 2004, IEEE Computer Society Press: Los Alamitos, CA, (2004).
16.
WolskiR., SpringN. and HayesJ., Predicting the CPU availability of time-shared unix systems on the computational grid, in: Proceedings of the 8th High-Performance Distributed Computing Conference, (1999).
17.
FosterI. and KesselmanC., Globus: A tookkit-based grid architecture, in: The Grid: Blueprint for a new computing infrastructure, Morgan Kaufmann, (1999), 259-278.
18.
ArpaciR., DusseauC.A., VahdatM.A., LiuT.L., AndersonT.E. and PattersonA.D., The interaction of parallel and sequential workloads on a network of workstations, in: Proceedings of SIGMETRICS'95, (1995), 267-278.
19.
ShihW., YangC. and TsengS., A performance-based parallel loop scheduling on grid environments, The Journal of Supercomputing41 (2007), 247-267.
20.
ShihW., YangC. and TsengS., Using a performance-based skeleton to implement divisible load applications on grid computing environments, Journal of Information Science and Engineering25 (2009), 59-81.
21.
YangC., ShihW. and TsengS., Dynamic partitioning of loop iterations on heterogeneous PC clusters, The Journal of Supercomputing44 (2008), 1-23.
22.
DindaP., The statistical properties of host load, Scientific Programming7(3-4) (1999).
23.
WolskiR., SpringN. and HayesJ., Predicting the CPU availability of time-shared unix systems, in: Peoceedings of 8th IEEE High Performance Distributed Computing Conference (HPDC8), (1999).
24.
RabenseifnerR., KonigesE.A., ProstJ. and HedgesR., The parallel effective I/O bandwidth benchmark: b_eff io, Message Passing Interface Developer's and User's Conference (MPIDC), (2000).
25.
MucciJ.P. and LondonK., The cachebench report, 1998.
26.
CurnowJ.H. and WichmannA.B., A synthetic benchmark, The Computer Journal19(1) (1976), 43-49.
27.
AburtoA., Flops.c version 2.0. ftp://ftp.nosc.mil/pub/aburto [October 2014].
28.
WeickerP.R., Dhrystone: A synthetic systems programming benchmark, Commun ACM, 27(10) (1984), 1013-1030.
29.
European CrossGrid Project: http://www.crossgrid.org [December 2014].
30.
GridBench: A tool for benchmarking Grids. http://grid.ucy.ac.cy/GridBench [December 2014].
31.
Tiramo-RamosA., TsouloupasG., DikaiakosD.M. and SlootP., Grid resource selection by application benchmarking: A computational haemodynamics case study, in: Proceedings of the International Conference on Computational Science Springer, (2005).
32.
TsouloupasG. and DikaiakosD.M., GridBench: A tool for benchmarking grids, in: Proceedings of the 4th International Workshop on Grid Computing (Grid2003), IEEE Computer Society, (2003), 60-67.
33.
TsouloupasG. and DikaiakosD.M., Characterization of computational grid resources using low-level benchmarks, Technical Report TR-2004-5, Dept. of Computer Science, University of Cyprus, December 2004.
34.
TsouloupasG. and DikaiakosD.M., Gridbench: A workbench for grid benchmarking, in: EGC, SlootP.M.A., HoekstraA.G., PriolT., ReinefeldA. and BubakM., eds, volume 3470 of Lecture Notes in Computer Science, Springer, 2005, 211-225.
35.
DikaiakosD.M., Grid benchmarking: vision, challenges, and current status, Concurrency Computat Pract Exper19 (2007), 89-105.
