sibsutisВестник СибГУТИThe Herald of the Siberian State University of Telecommunications and Information Science1998-6920СибГУТИsibsutis-460Research ArticleСтатьиАлгоритмы отказоустойчивого управления ресурсами пространственно-распределённых вычислительных системAlgorithms of fault-tolerant resources management of geographically distributed computer systemsПоляковА. Ю.PolyakovA. Y.artpol@csc.sibsutis.ruМолдовановаО. В.MoldovanovaO. V.ovm@csc.sibsutis.ruПазниковА. А.PaznikovA. A.apaznikov@gmail.comКурносовМ. Г.KernosovM. G.mkurnosov@gmail.commailto:apaznikov@gmail.comМамойленкоС. Н.MamoilenkoS. N.msn@sibsutis.ruЕфимовА. В.EfimovA. V.efimov@cpct.sibsutis.ruСибГУТИ; ИФП СО РАНRussian FederationСибГУТИRussian Federation201426102022041129Copyright © Поляков А.Ю., Молдованова О.В., Пазников А.А., Курносов М.Г., Мамойленко С.Н., Ефимов А.В., 20222022Поляков А.Ю., Молдованова О.В., Пазников А.А., Курносов М.Г., Мамойленко С.Н., Ефимов А.В.Polyakov A.Y., Moldovanova O.V., Paznikov A.A., Kernosov M.G., Mamoilenko S.N., Efimov A.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vestnik.sibsutis.ru/jour/article/view/460

В статье рассматриваются вопросы организации отказоустойчивого функционирования пространственно-распределённых вычислительных систем (ПРВС) в мультизадачных режимах. Предложена модель функционирования ПРВС, состоящих из множества подсистем. Разработаны алгоритмы функционирования распределённой очереди задач ПРВС. Для территориально-сосредоточенных вычислительных подсистем созданы алгоритмы управления ресурсами при обслуживании пользовательских задач, представленных параллельными масштабируемыми программами. Развит программный инструментарий отказоустойчивого выполнения программ на подсистемах ПРВС. Приведены результаты моделирования алгоритмов на мультикластерной ПРВС (GRID-модель).

This paper considers the issues of fault-tolerant operation of geographically distributed computer systems (GDCS) in multitask modes. The model of GDCS operation consisting of many subsystems was proposed. Algorithms of fault-tolerant distributed job queue operation were developed. Algorithms of resource management for the GDCS subsystems were created. The program tools of fault-tolerant execution of parallel programs on GDCS subsystems were developed and implemented. Simulation results of the algorithms on the multicluster GDCS (GRID model) were presented.

распределённые вычислительные системыпараллельное мульти-программированиемультикластерные системыдиспетчеризацияотказоустойчивостьMPIGRIDdistributed computer systemsparallel multiprogrammingMPIGRIDmulticluster systemsdispatchingfault-toleranceand self-diagnosisReferencesХорошевский В.Г. Распределённые вычислительные системы с программируемой структурой// Вестник СибГУТИ. 2010. №2 (10). С. 3-41.Хорошевский В.Г. Распределённые вычислительные системы с программируемой структурой// Вестник СибГУТИ. 2010. №2 (10). С. 3-41.Torque Resource Manager [Электронный ресурс]. Режим доступа: http://www.adaptivecomputing.com/products/open-source/torque/ (дата обращения 03.09.2014).Torque Resource Manager [Электронный ресурс]. Режим доступа: http://www.adaptivecomputing.com/products/open-source/torque/ (дата обращения 03.09.2014).SLURM: Simple Linux Utility for Resource Management, A. Yoo, M. Jette, and M. Grondona, Job Scheduling Strategies for Parallel Processing, volume 2862 of Lecture Notes in Computer Science, pages 44-60, Springer-Verlag, 2003.SLURM: Simple Linux Utility for Resource Management, A. Yoo, M. Jette, and M. Grondona, Job Scheduling Strategies for Parallel Processing, volume 2862 of Lecture Notes in Computer Science, pages 44-60, Springer-Verlag, 2003.Huedo E., Montero R.S., Llorente I.M. A framework for adaptive execution on grids // Software - Practice and Experience (SPE). 2004. Vol. 34. P. 631-651.Huedo E., Montero R.S., Llorente I.M. A framework for adaptive execution on grids // Software - Practice and Experience (SPE). 2004. Vol. 34. P. 631-651.Berman F., Wolski R., Casanova H. Adaptive computing on the grid using AppLeS // IEEE Trans. on Parallel and Distributed Systems. 2003. Vol. 34. P. 369-382.Berman F., Wolski R., Casanova H. Adaptive computing on the grid using AppLeS // IEEE Trans. on Parallel and Distributed Systems. 2003. Vol. 34. P. 369-382.Cooper K., Dasgupta A., Kennedy C.K. [et al]. New Grid Scheduling and Rescheduling Methods in the GrADS Project // Proc. of the 18th International Parallel and Distributed Processing Symposium (IPDPS’04). 2004. Vol. 34. P. 199-206.Cooper K., Dasgupta A., Kennedy C.K. [et al]. New Grid Scheduling and Rescheduling Methods in the GrADS Project // Proc. of the 18th International Parallel and Distributed Processing Symposium (IPDPS’04). 2004. Vol. 34. P. 199-206.Buyya R., Abramson D., Giddy J. Nimrod/G: An architecture for a resource management and scheduling system in a global computational Grid // Proc. of the 4th International Conference on High Performance Computing in Asia-Pacific Region. 2000. P. 283-289.Buyya R., Abramson D., Giddy J. Nimrod/G: An architecture for a resource management and scheduling system in a global computational Grid // Proc. of the 4th International Conference on High Performance Computing in Asia-Pacific Region. 2000. P. 283-289.Frey J., Tannenbaum T., Livny M. [et al.] Condor-G: A computation management agent for multi-institutional grids // Cluster Computing. 2001. Vol. 5. P. 237-246.Frey J., Tannenbaum T., Livny M. [et al.] Condor-G: A computation management agent for multi-institutional grids // Cluster Computing. 2001. Vol. 5. P. 237-246.Andreetto P., Borgia S., Dorigo A. Practical approaches to grid workload and resource management in the EGEE project / // In CHEP ’04: Proceedings of the Conference on Computing in High Energy and Nuclear Physics. 2004. Vol. 2. P. 899-902.Andreetto P., Borgia S., Dorigo A. Practical approaches to grid workload and resource management in the EGEE project / // In CHEP ’04: Proceedings of the Conference on Computing in High Energy and Nuclear Physics. 2004. Vol. 2. P. 899-902.Wijngaards N., Overeinder B., Steen M., Brazier F. Supporting internet-scale multi-agent systems // Data Knowledge Engineering. 2002. Vol. 41. P. 229-245.Wijngaards N., Overeinder B., Steen M., Brazier F. Supporting internet-scale multi-agent systems // Data Knowledge Engineering. 2002. Vol. 41. P. 229-245.Caron E., Garonne V., Tsaregorodtsev A. Evaluation of Meta-scheduler Architectures and Task Assignment Policies for High Throughput Computing // Technical report № 5576. Institut National de Recherche en Informatique et en Automatique. 2005. 16p.Caron E., Garonne V., Tsaregorodtsev A. Evaluation of Meta-scheduler Architectures and Task Assignment Policies for High Throughput Computing // Technical report № 5576. Institut National de Recherche en Informatique et en Automatique. 2005. 16p.Deelman E., Singh G., Su M.-H. [et al.] Pegasus: A framework for mapping complex scientific workflows onto distributed systems // Scientific Programming. 2005. Vol. 13(3). P. 219-237.Deelman E., Singh G., Su M.-H. [et al.] Pegasus: A framework for mapping complex scientific workflows onto distributed systems // Scientific Programming. 2005. Vol. 13(3). P. 219-237.Hull D., Wolstencroft K., Stevens R. [et al.] Taverna: a tool for building and running workflows of services / // Nucleic Acids Research. 2006. Vol. 34. P. 729-732.Hull D., Wolstencroft K., Stevens R. [et al.] Taverna: a tool for building and running workflows of services / // Nucleic Acids Research. 2006. Vol. 34. P. 729-732.Matthew I., Shields M., Wang I., Philp R. Grid Enabling Applications Using Triana // In Workshop on Grid Applications and Programming Tools. 2003. 11p.Matthew I., Shields M., Wang I., Philp R. Grid Enabling Applications Using Triana // In Workshop on Grid Applications and Programming Tools. 2003. 11p.Fahringer T., Prodan R., Duan R. [et al.] ASKALON: A Grid Application Development and Computing Environment // 6th IEEE/ACM International Workshop on Grid Computing. 2005. P. 122-131.Fahringer T., Prodan R., Duan R. [et al.] ASKALON: A Grid Application Development and Computing Environment // 6th IEEE/ACM International Workshop on Grid Computing. 2005. P. 122-131.Young L., Mcgough S., Newhouse S., Darlington J. Scheduling Architecture and Algorithms within the ICENI Grid Middleware // In UK e-Science All Hands Meeting. 2003. P. 5-12.Young L., Mcgough S., Newhouse S., Darlington J. Scheduling Architecture and Algorithms within the ICENI Grid Middleware // In UK e-Science All Hands Meeting. 2003. P. 5-12.Altintas I., Berkley C., Jaeger E. [et al.] Kepler: An Extensible System for Design and Execution of Scientific Workflows // International Conference on Scientific and Statistical Database Management. 2004. P. 21-23.Altintas I., Berkley C., Jaeger E. [et al.] Kepler: An Extensible System for Design and Execution of Scientific Workflows // International Conference on Scientific and Statistical Database Management. 2004. P. 21-23.Kurnosov M., Paznikov A. Efficiency analysis of decentralized grid scheduling with job migration and replication // ACM International Conference on Ubiquitous Information Management and Communication. 2013. 7 p.Kurnosov M., Paznikov A. Efficiency analysis of decentralized grid scheduling with job migration and replication // ACM International Conference on Ubiquitous Information Management and Communication. 2013. 7 p.Feitelson D.G., [et. al]. Theory and practice in parallel job scheduling // Job Scheduling Strategies for Parallel Processing. 1997. Vol. 1291. P. 1 - 34.Feitelson D.G., [et. al]. Theory and practice in parallel job scheduling // Job Scheduling Strategies for Parallel Processing. 1997. Vol. 1291. P. 1 - 34.Shmueli E.,Feitelson D.G. Backfilling with lookahead to optimize the packing of parallel jobs. J. Parallel & Distributed Comput. 2005. Vol. 65. Iss. 9. P. 1090 - 1107.Shmueli E.,Feitelson D.G. Backfilling with lookahead to optimize the packing of parallel jobs. J. Parallel & Distributed Comput. 2005. Vol. 65. Iss. 9. P. 1090 - 1107.Cirne W., Grande C., Berman F. When the herd is smart aggregate behavior in the selection of job request. IEEE Transactions in Parallel and Distributed Systems. 2003. Vol. 14. P. 181 - 192.Cirne W., Grande C., Berman F. When the herd is smart aggregate behavior in the selection of job request. IEEE Transactions in Parallel and Distributed Systems. 2003. Vol. 14. P. 181 - 192.Мамойленко С.Н., Ефимов А.В. Алгоритмы планирования решения масштабируемых задач на распределённых вычислительных системах. Вестник ГОУ ВПО «СибГУТИ». 2010. № 2. С. 66 - 78.Мамойленко С.Н., Ефимов А.В. Алгоритмы планирования решения масштабируемых задач на распределённых вычислительных системах. Вестник ГОУ ВПО «СибГУТИ». 2010. № 2. С. 66 - 78.Cirne W., Berman F. A model for moldable supercomputer jobs. 15th Intl. Parallel & Distributed Processing Symp. 2001 URL: http://cseweb.ucsd.edu/~walfredo/papers/ moldability-model.pdf (дата обращения: 29.09.2014).Cirne W., Berman F. A model for moldable supercomputer jobs. 15th Intl. Parallel & Distributed Processing Symp. 2001 URL: http://cseweb.ucsd.edu/~walfredo/papers/ moldability-model.pdf (дата обращения: 29.09.2014).Elnozahy, E.N., [et. al.] A survey of rollback-recovery protocols in message-passing systems. ACM Computing Surveys. 2002. Vol. 34. N. 3. P. 375 - 408.Elnozahy, E.N., [et. al.] A survey of rollback-recovery protocols in message-passing systems. ACM Computing Surveys. 2002. Vol. 34. N. 3. P. 375 - 408.Ansel J., Arya K., Cooperman G. DMTCP: Transparent Checkpointing for Cluster Computations and the Desktop. IEEE International Parallel and Distributed Processing Symposium (IPDPS'09). 2009. 12 p.Ansel J., Arya K., Cooperman G. DMTCP: Transparent Checkpointing for Cluster Computations and the Desktop. IEEE International Parallel and Distributed Processing Symposium (IPDPS'09). 2009. 12 p.Поляков А.Ю. О восстановлении программ из контрольной точки / А.Ю. Поляков. Вестник ЮУрГУ. Серия «Математическое моделирование и программирование». 2010. № 35(211). С. 91 - 103.Поляков А.Ю. О восстановлении программ из контрольной точки / А.Ю. Поляков. Вестник ЮУрГУ. Серия «Математическое моделирование и программирование». 2010. № 35(211). С. 91 - 103.

The authors declare that there are no conflicts of interest present.