The optimal resource allocation to satisfy such demands and the proper settlement of contention when demands exceed the capacity of the resources, constitute the problem of being able to understand and to predict system behavior. To this analysis we can use both analytical and simulation methods. Modeling and simulation are methods, which are commonly used by performance analysts to represent constraints and to optimize performance. Principally analytical methods represented first of all by queuing theory belongs to the preferred method in comparison to the simulation method, because of their potential ability of general analysis and also of their ability to potentially analyze also massive parallel computers. But these arguments supposed to develop and to verify suggested analytical models. This article goes further in applying the achieved analytical results in queuing theory for complex performance evaluation in parallel computing [9, 14]. The extensions are mainly in extending derived analytical models to whole range of parallel computers including massive parallel computers (Grid, meta computer). The article therefore describes standard analytical model based on M/M/m, M/D/m and M/M/1, M/D/1 queuing theory systems. Then the paper describes derivation of the correction factor for standard analytical model, based on M/M/m and M/M/1 queuing systems, to study more precise their basic performance parameters (overhead latencies, throughput etc.). All the derived analytical models were compared with performed simulation results in order to estimate the magnitude of improvement. Likewise they were tested under various ranges of parameters, which influence the architecture of the parallel computers and its communication networks too. These results are very important in practical use.
| Published in |
American Journal of Networks and Communications (Volume 3, Issue 5-1)
This article belongs to the Special Issue Parallel Computer and Parallel Algorithms |
| DOI | 10.11648/j.ajnc.s.2014030501.14 |
| Page(s) | 43-56 |
| Creative Commons |
This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited. |
| Copyright |
Copyright © The Author(s), 2014. Published by Science Publishing Group |
Parallel Computer, NOW, Grid, Communication System, Jackson Theorem, Correction Factor, Analytical model, Performance Modeling, Queuing System
| [1] | Abderazek A. B., Multicore systems on chip - Practical Software/Hardware design, Imperial college press, United Kingdom, pp. 200, 2010 |
| [2] | Arora S., Barak B., Computational complexity - A modern Approach, Cambridge University Press, UK, pp. 573, 2009 |
| [3] | Dattatreya G. R., Performance analysis of queuing and computer network, University of Texas, Dallas, USA, pp. 472, 2008 |
| [4] | Dubois M., Annavaram M., Stenstrom P., Parallel Computer Organization and Design, Cambridge university press, United Kingdom, pp. 560, 2012 |
| [5] | Dubhash D.P., Panconesi A., Concentration of measure for the analysis of randomized algorithms, Cambridge University Press, UK, 2009 |
| [6] | Gelenbe E., Analysis and synthesis of computer systems, Imperial College Press, UK; pp. 324, 2010 |
| [7] | Giambene G., Queuing theory and telecommunications, Springer, pp. 585, 2005 |
| [8] | Hager G., Wellein G., Introduction to High Performance Computing for Scientists and Engineers, CRC Press, USA, pp. 356, 2010 |
| [9] | Hanuliak Peter, Hanuliak Michal, Modeling of single computing nodes of parallel computers, American J. of Networks and Communication, Science PG, Vol.3, USA, 2014 |
| [10] | Hanuliak J., Hanuliak I., To performance evaluation of distributed parallel algorithms, Kybernetes, Volume 34, No. 9/10, UK, pp. 1633-1650, 2005 |
| [11] | Hanuliak P., Analytical method of performance prediction in parallel algorithms, The Open Cybernetics and Systemics Journal, Vol. 6, Bentham, UK, pp. 38-47,2012 |
| [12] | Hanuliak P., Hanuliak I., Performance evaluation of iterative parallel algorithms, Kybernetes, Volume 39, No.1/ 2010, United Kingdom, pp. 107- 126, 2010 |
| [13] | Hanuliak J., Modeling of communication complexity in parallel computing, American J. of Networks and Communication, Science PG, Vol. 3, USA, 2014 |
| [14] | Hanuliak P., Complex analytical performance modeling of parallel algorithms, American J. of Networks and Communication, Science PG, Vol. 3, USA, 2014 |
| [15] | Hanuliak P., Hanuliak J., Complex performance modeling of parallel algorithms, American J. of Networks and Communication, Science PG, Vol. 3, USA, 2014 |
| [16] | Harchol-BalterMor, Performance modeling and design of computer systems, Cambridge University Press, United Kingdom, pp. 576, 2013 |
| [17] | Hillston J., A Compositional Approach to Performance Modeling, University of Edinburg, Cambridge University Press, United Kingdom, pp. 172, 2005 |
| [18] | Hwang K. and coll., Distributed and Parallel Computing, Morgan Kaufmann, USA, pp. 472, 2011 |
| [19] | Kshemkalyani A. D., Singhal M., Distributed Computing, University of Illinois, Cambridge University Press, United Kingdom , pp. 756, 2011 |
| [20] | Kirk D. B., Hwu W. W., Programming massively parallel processors, Morgan Kaufmann, USA, pp. 280, 2010 |
| [21] | Kostin A., Ilushechkina L., Modeling and simulation of distributed systems, Imperial College Press, United Kingdom, pp. 440, 2010 |
| [22] | Kumar A., Manjunath D., Kuri J., Communication Networking , Morgan Kaufmann, USA, pp. 750, 2004 |
| [23] | Kushilevitz E., Nissan N., Communication Complexity, Cambridge University Press, United Kingdom, pp. 208, 2006 |
| [24] | Kwiatkowska M., Norman G., and Parker D., PRISM 4.0: Verification of Probabilistic Real-time Systems, In Proc. of 23rd CAV’11, Vol. 6806, Springer, Germany, pp. 585-591, 2011 |
| [25] | Le Boudec Jean-Yves, Performance evaluation of computer and communication systems, CRC Press, USA, pp. 300, 2011 |
| [26] | McCabe J., D., Network analysis, architecture, and design (3rd edition), Elsevier/ Morgan Kaufmann, USA, pp. 496, 2010 |
| [27] | Meerschaert M., Mathematical modeling (4-th edition), Elsevier, Netherland, pp. 384, 2013 |
| [28] | Misra Ch. S.,Woungang I., Selected topics in communication network and distributed systems, Imperial college press, UK, pp. 808, 2010 |
| [29] | Natarajan Gautam, Analysis of Queues: Methods and Applications, CRC Press , USA, pp. 802, 2012 |
| [30] | Peterson L. L., Davie B. C., Computer networks – a systemapproach, Morgan Kaufmann, USA, pp. 920, 2011 |
| [31] | Resch M. M., Supercomputers in Grids, Int. J. of Grid and HPC, No.1, pp. 1 - 9, 2009 |
| [32] | Riano l., McGinity T.M., Quantifying the role of complexity in a system´s performance, Evolving Systems, Springer Verlag, Germany, pp. 189 – 198, 2011 |
| [33] | Ross S. M., Introduction to Probability Models, 10th edition, Academic Press, Elsevier Science, Netherland, pp. 800, 2010 |
| [34] | Wang L., Jie Wei., Chen J., Grid Computing: Infrastructure, Service, and Application, CRC Press, USA, 2009www pages |
| [35] | www.top500.org. |
APA Style
Michal Hanuliak. (2014). Modeling of Parallel Computers Based on Network of Computing. American Journal of Networks and Communications, 3(5-1), 43-56. https://doi.org/10.11648/j.ajnc.s.2014030501.14
ACS Style
Michal Hanuliak. Modeling of Parallel Computers Based on Network of Computing. Am. J. Netw. Commun. 2014, 3(5-1), 43-56. doi: 10.11648/j.ajnc.s.2014030501.14
AMA Style
Michal Hanuliak. Modeling of Parallel Computers Based on Network of Computing. Am J Netw Commun. 2014;3(5-1):43-56. doi: 10.11648/j.ajnc.s.2014030501.14
Share This Article
Twitter
Linked In
Facebook
Copy
Download@article{10.11648/j.ajnc.s.2014030501.14,
author = {Michal Hanuliak},
title = {Modeling of Parallel Computers Based on Network of Computing},
journal = {American Journal of Networks and Communications},
volume = {3},
number = {5-1},
pages = {43-56},
doi = {10.11648/j.ajnc.s.2014030501.14},
url = {https://doi.org/10.11648/j.ajnc.s.2014030501.14},
eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ajnc.s.2014030501.14},
abstract = {The optimal resource allocation to satisfy such demands and the proper settlement of contention when demands exceed the capacity of the resources, constitute the problem of being able to understand and to predict system behavior. To this analysis we can use both analytical and simulation methods. Modeling and simulation are methods, which are commonly used by performance analysts to represent constraints and to optimize performance. Principally analytical methods represented first of all by queuing theory belongs to the preferred method in comparison to the simulation method, because of their potential ability of general analysis and also of their ability to potentially analyze also massive parallel computers. But these arguments supposed to develop and to verify suggested analytical models. This article goes further in applying the achieved analytical results in queuing theory for complex performance evaluation in parallel computing [9, 14]. The extensions are mainly in extending derived analytical models to whole range of parallel computers including massive parallel computers (Grid, meta computer). The article therefore describes standard analytical model based on M/M/m, M/D/m and M/M/1, M/D/1 queuing theory systems. Then the paper describes derivation of the correction factor for standard analytical model, based on M/M/m and M/M/1 queuing systems, to study more precise their basic performance parameters (overhead latencies, throughput etc.). All the derived analytical models were compared with performed simulation results in order to estimate the magnitude of improvement. Likewise they were tested under various ranges of parameters, which influence the architecture of the parallel computers and its communication networks too. These results are very important in practical use.},
year = {2014}
}
TY - JOUR T1 - Modeling of Parallel Computers Based on Network of Computing AU - Michal Hanuliak Y1 - 2014/07/31 PY - 2014 N1 - https://doi.org/10.11648/j.ajnc.s.2014030501.14 DO - 10.11648/j.ajnc.s.2014030501.14 T2 - American Journal of Networks and Communications JF - American Journal of Networks and Communications JO - American Journal of Networks and Communications SP - 43 EP - 56 PB - Science Publishing Group SN - 2326-8964 UR - https://doi.org/10.11648/j.ajnc.s.2014030501.14 AB - The optimal resource allocation to satisfy such demands and the proper settlement of contention when demands exceed the capacity of the resources, constitute the problem of being able to understand and to predict system behavior. To this analysis we can use both analytical and simulation methods. Modeling and simulation are methods, which are commonly used by performance analysts to represent constraints and to optimize performance. Principally analytical methods represented first of all by queuing theory belongs to the preferred method in comparison to the simulation method, because of their potential ability of general analysis and also of their ability to potentially analyze also massive parallel computers. But these arguments supposed to develop and to verify suggested analytical models. This article goes further in applying the achieved analytical results in queuing theory for complex performance evaluation in parallel computing [9, 14]. The extensions are mainly in extending derived analytical models to whole range of parallel computers including massive parallel computers (Grid, meta computer). The article therefore describes standard analytical model based on M/M/m, M/D/m and M/M/1, M/D/1 queuing theory systems. Then the paper describes derivation of the correction factor for standard analytical model, based on M/M/m and M/M/1 queuing systems, to study more precise their basic performance parameters (overhead latencies, throughput etc.). All the derived analytical models were compared with performed simulation results in order to estimate the magnitude of improvement. Likewise they were tested under various ranges of parameters, which influence the architecture of the parallel computers and its communication networks too. These results are very important in practical use. VL - 3 IS - 5-1 ER -
Email: