A Resource and Job Management System (RJMS) is a crucial system software part of the HPC stack. It is responsible for efficiently delivering computing power to applications in supercomputing environments. Its main intelligence relies on resource selection techniques to find the most adapted resources to schedule the users’ jobs. This article introduces a new method that takes into account the topology of the machine and the application characteristics to determine the best choice among the available nodes of the platform, based upon the network topology and taking into account the application communication pattern. To validate our approach, we integrate this algorithm as a plugin for Simple Linux Utility for Resource Management (SLURM), a well-known and widespread RJMS. We assess our plugin with different optimization schemes by comparing with the default topology-aware Slurm algorithm, using both emulation and simulation of a large-scale platform and by carrying out experiments in a real cluster. We show that transparently taking into account a job communication pattern and the topology allows for relevant performance gains.
AlbingCTroullierNWhalenS. (2011) Scalable node allocation for improved performance in regular and anisotropic 3d torus supercomputers. In: EuroMPI 2011, Santorini, Greece, 18–21 September 2011, pp. 61–70. Springer.
2.
BalleSMPalermoDJ (2007) Enhancing an open source resource manager with multi-core/multi-threaded support. In: JSSPP Job Scheduling Strategies for Parallel Processing 2007, Seattle, WA, USA, 17 June 2007, pp. 37–50. Springer.
3.
BarrettRFVaughanCTHerouxMA (2011) Minighost: A Miniapp for exploring boundary exchange strategies using stencil computations in scientific parallel computing. In: Sandia National Laboratories, Tech. Rep. SAND2011-5294832.
4.
BhateléABohmEJKaléLV (2009) Topology aware task mapping techniques: An API and case study. In: Proceedings of the 14th ACM SIGPLAN symposium on Principles and practice of parallel programming (PPoPP '09), pp. 301–302. New York, NY: ACM.
5.
BosilcaGFoyerCJeannotE. (2017) Online dynamic monitoring of MPI communication. In: 23rd International European Conference on Parallel and Distributed Computing (EuroPar), Santiago de Compostella, 28 August–1 September, p. 12. Springer.
6.
BroquedisFClet-OrtegaJMoreaudS. (2010) Hwloc: A generic framework for managing hardware affinities in HPC applications. In: Proceedings of the 2010 18th Euromicro Conference on Parallel, Distributed and Network-based Processing (PDP '10), pp. 180–186. Washington, DC: IEEE Computer Society.
7.
CapitNDa CostaGGeorgiouY. (2005) A batch scheduler with high level components. In: Proceedings of the Fifth IEEE International Symposium on Cluster Computing and the Grid (CCGrid'05)–Volume 2–02 (CCGRID '05), Vol. 2. pp. 776–783. Washington, DC: IEEE Computer Society.
8.
CruzEHDienerMPillaLL. (2015) An efficient algorithm for communication-based task mapping. In: Proceedings of the 2015 23rd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing (PDP '15), pp. 207–214. Washington, DC: IEEE Computer Society.
GeorgiouYHautreuxM (2012) Evaluating scalability and efficiency of the resource and job management system on large HPC clusters. In: Job Scheduling Strategies for Parallel Processing: 16th International Workshop, JSSPP 2012, Shanghai, China, 25 May 2013, pp. 134–156. Berlin, Heidelberg: Springer.
11.
HoeflerTSnirM (2011) Generic topology mapping strategies for large-scale parallel architectures. In: Proceedings of the International Conference on Supercomputing (ICS '11), pp. 75–84. New York, NY: ACM.
12.
JeannotEMercierG (2010) Near-optimal placement of MPI processes on hierarchical Numa architectures. In: D’AmbraPGuarracinoMTaliaD (eds) Proceedings of the 16th International Euro-Par Conference on Parallel Processing: Part II (Euro-Par'10), pp. 199–210. Berlin, Heidelberg: Springer-Verlag.
13.
JeannotEMercierGTessierF (2014) Process placement in multicore clusters: Algorithmic issues and practical techniques. IEEE Trans Parallel Distrib Syst25(4): 993–1002.
14.
LeeGToliaNRanganathanP. (2010) Topology-aware resource allocation for data-intensive workloads. In: SIGCOMM Computer Communication Review41(1): 120–124.
15.
LiuCYangLFosterI. (2002) Design and evaluation of a resource selection framework for grid applications. In: Proceedings 11th IEEE International Symposium on High Performance Distributed Computing, 2002, pp. 63–72.
16.
MercierGClet-OrtegaJ (2009) Towards an efficient process placement policy for MPI applications in multicore environments. In: RopoMWesterholmJDongarraJ (eds) Proceedings of the 16th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface, pp. 104–115. Berlin, Heidelberg: Springer-Verlag.
17.
MercierGJeannotE (2011) Improving MPI applications performance on multicore clusters with rank reordering. In: CotronisYDanalisANikolopoulosDS. (eds) Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface (EuroMPI'11), pp. 39–49. Berlin, Heidelberg: Springer-Verlag.
RamanRLivnyMSolomonM (1998) Matchmaking: Distributed resource management for high throughput computing. In: Proceedings. The Seventh International Symposium on High Performance Distributed Computing (Cat. No.98TB100244), Chicago, IL, 1998, pp. 140–146.
22.
RashtiMJGreenJBalajiP. (2011) Multi-core and network aware MPI topology functions. In: CotronisYDanalisANikolopoulosDS. (eds) Proceedings of the 18th European MPI Users' Group conference on Recent advances in the message passing interface (EuroMPI'11), pp. 50–60. Berlin, Heidelberg: Springer-Verlag.
23.
SantosCASahaiAZhuX. (2004) Policy-based resource assignment in utility computing environments. In: Utility Computing: 15th IFIP/IEEE International Workshop on Distributed Systems: Operations and Management, DSOM 2004, Davis, CA, USA, 15–17 November 2004.
24.
SmithCMcMillanBLumbI (2001) Topology aware scheduling in the LSF distributed resource manager. In: Proceedings of the Cray User Group Meeting.
25.
SonmezOMohamedHEpemaD (2006) Communication-aware job placement policies for the Koala grid scheduler. In: 2006 Second IEEE International Conference on e-Science and Grid Computing (e-Science'06), 2006, pp. 79–79. Amsterdam: The Netherlands.
WuJXiongXLanZ (2015) Hierarchical task mapping for parallel applications on supercomputers. The Journal of Supercomputing71(5): 1776–1802.
28.
YangXZhouZTangW. (2014) Balancing job performance with system performance via locality-aware scheduling on torus-connected systems. In: 2014 IEEE International Conference on Cluster Computing (CLUSTER), Madrid, 2014, pp. 140–148.
29.
YooAJetteMGrondonaM (2003) Slurm: Simple Linux utility for resource management. In: Job Scheduling Strategies for Parallel Processing: 9th International Workshop, JSSPP 2003, Seattle, WA, USA, 24 June 2003. Berlin Heidelberg: Springer. pp. 44–60.
30.
YuHChungIHMoreiraJ (2006) Topology mapping for blue Gene/l supercomputer. In: Proceedings of the 2006 ACM/IEEE conference on Supercomputing (SC '06). New York: ACM.
