Abstract
In this paper we study a simultaneous service multiserver system which we call speed-scaling supercomputer, where speed-scaling is used to address the performance/power demand tradeoff. We treat the system by three-level modeling approach, using matrix-analytic method, generalized semi-Markov processes and small-scale technical system as the three levels of modeling. An explicit form of stability condition is obtained for a two-server system with heterogeneous customer classes. Regenerative estimation approach is used for confidence estimation of performance measures both in simulation and technical models. We demonstrate the potential of the three-level modeling approach on a relatively sophisticated and interesting model by performing extensive experiments.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
References
Afanaseva, L., Bashtova, E., & Grishunina, S. (2019). Stability analysis of a multi-server model with simultaneous service and a regenerative input flow. Methodology and Computing in Applied Probability. https://doi.org/10.1007/s11009-019-09721-9.
Armistead, R., & Li, Fangxing (2011). Parallel computing of sparse linear systems using matrix condensation algorithm. In 2011 IEEE Trondheim PowerTech (pp. 1–6). IEEE. https://doi.org/10.1109/PTC.2011.6238219. Retrieved May 31, 2022, from http://ieeexplore.ieee.org/document/6238219/
Arthurs, E., & Kaufman, J. S. (1979). Sizing a message store subject to blocking criteria. In Proceedings of the third international symposium on modelling and performance evaluation of computer systems: Performance of computer systems (pp. 547–564). North-Holland Publishing Co. http://dl.acm.org/citation.cfm?id=647409.724583
Asmussen, S. (2003). Applied probability and queues. In Stochastic modelling and applied probability (Vol. 51). Springer. https://doi.org/10.1007/b97236
Basmadjian, R. (2019). Flexibility-based energy and demand management in data centers: A case study for cloud computing. Energies. https://doi.org/10.3390/en12173301.
Basmadjian, R., Bouvry, P., Costa, G. D., Gyarmati, L., Kliazovich, D., Lafond, S., et al. (2015). Green data centers. In J.-M. Pierson (Ed.), Large-scale distributed systems and energy efficiency (pp. 159–196). Wiley. https://doi.org/10.1002/9781118981122.ch6.
Basmadjian, R., & de Meer, H. (2012). Evaluating and modeling power consumption of multi-core processors. In 2012 Third international conference on future systems: Where energy, computing and communication meet (e-Energy) (pp. 1–10). https://doi.org/10.1145/2208828.2208840
Basmadjian, R., & de Meer, H. (2018). Modelling and analysing conservative governor of DVFs-enabled processors. In Proceedings of the ninth international conference on future energy systems. e-Energy ’18, (pp. 519–525). Association for Computing Machinery. https://doi.org/10.1145/3208903.3213778
Basmadjian, R., Niedermeier, F., & de Meer, H. (2016). Modelling performance and power consumption of utilisation-based DVFS using m/m/1 queues. In Proceedings of the seventh international conference on future energy systems. e-Energy ’16. Association for Computing Machinery. https://doi.org/10.1145/2934328.2934342
Basmadjian, R., Rainer, S., & Meer, H. D. (2013). A generic methodology to derive empirical power consumption prediction models for multi-core processors. In 2013 international conference on cloud and green computing (pp. 167–174). https://doi.org/10.1109/CGC.2013.32
Beslin, S. J. (1992). Cofactor matrices. Linear Algebra and its Applications, 165, 45–52. https://doi.org/10.1016/0024-3795(92)90227-2.
Brill, P. H., & Green, L. (1984). Queues in which customers receive simultaneous service from a random number of servers: A system point approach. Management Science, 30(1), 51–68. https://doi.org/10.1287/mnsc.30.1.51.
Chakravarthy, S. R., & Karatza, H. D. (2013). Two-server parallel system with pure space sharing and Markovian arrivals. Computers & Operations Research, 40(1), 510–519. https://doi.org/10.1016/j.cor.2012.08.002.
Criado, J., Garcia-Gasulla, M., Kumbhar, P., Awile, O., Magkanaris, I., & Mantovani, F. (2020). Coreneuron: Performance and energy efficiency evaluation on intel and arm cpus. In 2020 IEEE international conference on cluster computing (CLUSTER) (pp. 540–548). https://doi.org/10.1109/CLUSTER49012.2020.00077
Evans, R. V. (1964). Queuing when jobs require several services which need not be sequenced. Management Science, 10(2), 298–315. https://doi.org/10.1287/mnsc.10.2.298.
Feitelson, D. G. (2015). Workload modeling for computer systems performance evaluation. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781139939690.
Filippopoulos, D., & Karatza, H. (2007). An M/M/2 parallel system model with pure space sharing among rigid jobs. Mathematical and Computer Modelling, 45(5–6), 491–530. https://doi.org/10.1016/j.mcm.2006.06.007.
Gail, H. R., Hantler, S. L., & Taylor, B. A. (1996). Spectral analysis of M/G/1 and G/M/1 type Markov chains. Advances in Applied Probability, 28(1), 114. https://doi.org/10.2307/1427915.
Gandhi, A., Gupta, V., Harchol-Balter, M., & Kozuch, M. A. (2010). Optimality analysis of energy-performance trade-off for server farm management. Performance 2010. Performance Evaluation, 67(11), 1155–1171. https://doi.org/10.1016/j.peva.2010.08.009.
Garimella, R. M., & Alexander, R. (2018). On an exact solution of the rate matrix of G/M/1-type Markov process with small number of phases. Journal of Parallel and Distributed Computing, 119, 172–178. https://doi.org/10.1016/j.jpdc.2018.04.013.
Glynn, P. W. (1989). A GSMP formalism for discrete event systems. Proceedings of the IEEE, 77(1), 14–23. https://doi.org/10.1109/5.21067.
Glynn, P. W. (1994). Some topics in regenerative steady-state simulation. Acta Applicandae Mathematicae, 34(1–2), 225–236. https://doi.org/10.1007/BF00994267.
Glynn, P. W., & Haas, P. J. (2006). Laws of large numbers and functional central limit theorems for generalized semi-Markov processes. Stochastic Models, 22(2), 201–231.
Glynn, P. W., & Haas, P. J. (2015). On transience and recurrence in irreducible finite-state stochastic systems. ACM Transactions on Modeling and Computer Simulation, 25(4), 1–19. https://doi.org/10.1145/2699721.
Glynn, P. W., & Heidelberger, P. (1992). Experiments with initial transient deletion for parallel. Replicated steady-state simulations. Management Science, 38(3), 400–418. https://doi.org/10.1287/mnsc.38.3.400.
Glynn, P. W., & Iglehart, D. L. (1988). Simulation methods for queues: An overview. Queueing Systems, 3(3), 221–255. https://doi.org/10.1007/BF01161216.
Grosof, I., Harchol-Balter, M., & Scheller-Wolf, A. (2020). Stability for two-class multiserver-job systems. arXiv:2010.00631 [cs]
Harchol-Balter, M. (2021). Open problems in queueing theory inspired by datacenter computing. Queueing Systems. https://doi.org/10.1007/s11134-020-09684-6.
Harchol-Balter, M. (2022). The multiserver job queueing model. Queueing Systems. https://doi.org/10.1007/s11134-022-09762-x.
Harrison, P. G., Patel, N. M., & Knottenbelt, W. J. (2016). Energy-performance trade-offs via the ep queue. ACM Transactions on Modeling and Performance Evaluation of Computing Systems, 1(2), 6–1631. https://doi.org/10.1145/2818726.
He, Q.-M. (2014). Fundamentals of matrix-analytic methods. Springer. https://doi.org/10.1007/978-1-4614-7330-5.
Henderson, S. G., & Glynn, P. W. (2001). Regenerative steady-state simulation of discrete-event systems. ACM Transactions on Modeling and Computer Simulation, 11(4), 313–345. https://doi.org/10.1145/508366.508367.
Hill, R. D., & Underwood, E. E. (1985). On the matrix adjoint (Adjugate). SIAM Journal on Algebraic Discrete Methods, 6(4), 731–737. https://doi.org/10.1137/0606071.
Hillier, F. S., & Lieberman, G. J. (2015). Introduction to operations research (10th ed.). McGraw-Hill.
Hyytiä, E., Down, D., Lassila, P., & Aalto, S. (2018). Dynamic control of running servers. In R. German, K.-S. Hielscher, & U. R. Krieger (Eds.), Measurement. Modelling and evaluation of computing systems. Lecture notes in computer science (Vol. 10740, pp. 127–141). Springer. https://doi.org/10.1007/978-3-319-74947-1_9
Jennings, O. B., Massey, W. A., & McCalla, C. (1997). Optimal profit for leased lines services (Vol. 2, pp. 803–814).
Kaufman, J. (1981). Blocking in a shared resource environment. IEEE Transactions on Communications, 29(10), 1474–1481. https://doi.org/10.1109/TCOM.1981.1094894.
Kaufman, J. (1992). Blocking with retrials in a completely shared resource environment. Journal of Performance Evaluation, 15(6), 99–113.
Kim, S. S. (1979). M-m-s queueing system where customers demand multiple server use. Ph.D. thesis, Southern Methodist University.
Kipnis, C., & Robert, P. (1990). A dynamic storage process. Stochastic Processes and Their Applications, 34(1), 155–169. https://doi.org/10.1016/0304-4149(90)90061-V.
Ko, Y. M., & Cho, Y. (2014). A distributed speed scaling and load balancing algorithm for energy efficient data centers. Performance Evaluation, 79, 120–133. https://doi.org/10.1016/j.peva.2014.07.008 (Special Issue: Performance 2014).
Kuehn, P. J., & Mashaly, M. (2019). DVFS-power management and performance engineering of data center server clusters. In 2019 15th annual conference on wireless on-demand network systems and services (WONS) (pp. 91–98). https://doi.org/10.23919/WONS.2019.8795470
Latouche, G., & Ramaswami, V. (1999). Introduction to matrix analytic methods in stochastic modeling. Society for Industrial and Applied Mathematics. https://doi.org/10.1137/1.9780898719734
Mandjes, M., & Sollie, B. (2021). A numerical approach for evaluating the time-dependent distribution of a quasi birth-death process. Methodology and Computing in Applied Probability. https://doi.org/10.1007/s11009-021-09882-6.
Mantovani, F., Garcia-Gasulla, M., Gracia, J., Stafford, E., Banchelli, F., Josep-Fabrego, M., et al. (2020). Performance and energy consumption of HPC workloads on a cluster based on arm thunderx2 CPU. Future Generation Computer Systems, 112, 800–818. https://doi.org/10.1016/j.future.2020.06.033.
Melikov, A. Z. (1996). Computation and optimization methods for multiresource queues. Cybernetics and Systems Analysis, 32(6), 821–836. https://doi.org/10.1007/BF02366862.
Morozov, E., & Rumyantsev, A. (2016). Stability analysis of a MAP/M/s cluster model by matrix-analytic method. In D. Fiems, M. Paolieri, & N. A. Platis (Eds.), Computer performance engineering: 13th European workshop, EPEW 2016, Chios, Greece, October 5–7, 2016, proceedings (pp. 63–76). Springer. https://doi.org/10.1007/978-3-319-46433-6_5
Nekrasova, R., & Rumyantsev, A. (2020a). Regenerative estimation of a simultaneous service multiserver system with speed scaling. In 2020 26th conference of open innovations association (FRUCT). Journal Abbreviation: 2020 26th conference of open innovations association (FRUCT) (pp. 346–351). https://doi.org/10.23919/FRUCT48808.2020.9087351
Nekrasova, R. S., & Rumyantsev, A. S. (2020b). Stability analysis and regenerative estimation of a simultaneous service system with speed scaling. Journal of Mathematical Sciences. https://doi.org/10.13140/RG.2.2.27875.68643 (to appear).
Neuts, M. F. (1981). Matrix-geometric solutions in stochastic models. Johns Hopkins University Press.
Plemmons, R. J. (1977). M-matrix characterizations. I-nonsingular M-matrices. Linear Algebra and its Applications, 18(2), 175–188. https://doi.org/10.1016/0024-3795(77)90073-8.
R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/
Rama Murthy, G., & Rumyantsev, A. (2018). On an exact solution of the rate matrix of g/m/1-type Markov process with small number of phases. Journal of Parallel and Distributed Computing, 119, 172–178. https://doi.org/10.1016/j.jpdc.2018.04.013.
Rumyantsev, A. (2020). Stability of multiclass multiserver models with automata-type phase transitions. In Proceedings of the second international workshop on stochastic modeling and applied research of technology (SMARTY 2020) (Vol. 2792, pp. 213–225). ISSN: 1613-0073. http://ceur-ws.org/Vol-2792/#paper16
Rumyantsev, A., Astafiev, S., & Golovin, A. (2021). R-Forge: Simulato: Project Home. https://r-forge.r-project.org/projects/simulato/
Rumyantsev, A., Basmadjian, R., Golovin, A., & Astafiev, S. (2021). A three-level modelling approach for asynchronous speed scaling in high-performance data centres. In Proceedings of the twelfth ACM international conference on future energy systems. e-Energy ’21 (pp. 417–423). Association for Computing Machinery. https://doi.org/10.1145/3447555.3466580
Rumyantsev, A., & Morozov, E. (2017). Stability criterion of a multiserver model with simultaneous service. Annals of Operations Research, 252(1), 29–39. https://doi.org/10.1007/s10479-015-1917-2.
Rumyantsev, A., Zueva, P., Kalinina, K., & Golovin, A. (2018). Evaluating a single-server queue with asynchronous speed scaling. In R. German, K.-S. Hielscher, & U. R. Krieger (Eds.), Measurement, modelling and evaluation of computing systems. Lecture notes in computer science (Vol. 10740, pp. 157–172). Springer. https://doi.org/10.1007/978-3-319-74947-1_11
Shajin, D., Krishnamoorthy, A., & Melikov, A. Z. (2019). On partial and complete blocking set of states in queueing-inventory model. Applied and Computational Mathematics, 18(2), 189–201.
Tikhonenko, O. M. (2005). Generalized Erlang problem for service systems with finite total capacity. Problems of Information Transmission, 41(3), 243–253. https://doi.org/10.1007/s11122-005-0029-z.
Ufuoma, O. (2013). A new and simple method of solving large linear systems: Based on Cramer’s rule but employing Dodgson’s condensation. In Lecture notes in engineering and computer science (Vol. 1, pp. 123–128). https://www.scopus.com/inward/record.uri?eid=2-s2.0-84903484388 &partnerID=40 &md5=32d4d0ab101a562b9a4404d6e76f1155
Unwin, A. R. (1984). Results for dual resource queues. In F. Baccelli & G. Fayolle (Eds.), Modelling and performance evaluation methodology. Lecture notes in control and information sciences (Vol. 60, pp. 351–370). Springer. https://doi.org/10.1007/BFb0005182. Retrieved May 30, 2022, from http://link.springer.com/10.1007/BFb0005182
von Kistowski, J., Block, H., Beckett, J., Lange, K.-D., Arnold, J. A., & Kounev, S. (2015). Analysis of the influences on server power consumption and energy efficiency for CPU-intensive workloads. In Proceedings of the 6th ACM/SPEC international conference on performance engineering. ICPE’15 (pp. 223–234). ACM. https://doi.org/10.1145/2668930.2688057
Wang, W., Xie, Q., & Harchol-Balter, M. (2021). Zero queueing for multi-server jobs. Proceedings of the ACM on Measurement and Analysis of Computing Systems, 5(1), 1–25. https://doi.org/10.1145/3447385.
Acknowledgements
The authors would like to thank the Editors of Special Issue and the referees for comments that helped to improve the paper. The publication has been prepared with the support of Russian Science Foundation according to the research Project No. 21-71-10135, https://rscf.ru/en/project/21-71-10135/. Authors thank Prof. Srinivas Chakravarthy and Prof. Garimella Rama Murthy for helpful discussions.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors have no relevant financial or non-financial interests to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Rumyantsev, A., Basmadjian, R., Astafiev, S. et al. Three-level modeling of a speed-scaling supercomputer. Ann Oper Res 331, 649–677 (2023). https://doi.org/10.1007/s10479-022-04830-0
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10479-022-04830-0