Abstract
Applications running on large scale systems often suffer from degraded performance and lack of reproducible run-times due to network-level congestion, whether caused by the application network traffic itself, or by unrelated background network traffic (i.e. other applications). This paper describes the hardware-based congestion control algorithm implemented in NVIDIA’s Quantum HDR 200 Gb/s InfiniBand generation and the AI-based training used to obtain algorithm parameters. The hardware leverages NVIDIA’s Data Center Quantized Congestion Notification (DCQCN) algorithm and protocol and applies it to the InfiniBand network layer. Congestion patterns described in the literature are studied and enhanced to create greater congestion and are used to study the impact of such patterns on three applications: Incompact3D, LAMMPS and VASP. The study shows that network congestion increases individual measured application run time by up to a factor of ten or greater, while introduction of the implemented congestion control on the Quantum HDR InfiniBand technology recovers most of the lost time for the tested applications and congestion.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
C. Zimmer, S. Atchley, R. Pankajakshan, et al.: An evaluation of the CORAL interconnects. In: Proceedings of the International Conference for High Performance Computing, pp. 1–18 (2019). https://doi.org/10.1145/3295500.3356166
Geoffray, P., Hoefler, T.: Adaptive routing strategies for modern high performance networks. In: 16th IEEE Symposium on High Performance Interconnects (Hot Interconnects), pp. 165–172 (2008). https://doi.org/10.1109/HOTI.2008.21
Mittal, R., et al.: Revisiting network support for RDMA. In: Proceedings of the 2018 Conference of the ACM Special Interest Group on Data Communication, pp. 313–326 (2018) https://doi.org/10.1145/3230543.3230557
Chunduri, S., Groves, T., Mendygral, P., et al.: GPCNeT: designing a benchmark suite for inducing and measuring contention in HPC networks. In: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC 2019), pp. 1–33 (2019). https://doi.org/10.1145/3295500.3356215
Clos, C.: A study of nonblocking switching networks. Bell Syst. Technol. J. 32(2), 406–424 (1953). https://doi.org/10.1002/j.1538-7305.1953.tb01433.x
Ngai, J., Seitz, C.: A framework for adaptive routing in multicomputer networks. In: Proceedings of ACM Symposium on Parallel Algorithms and Architectures (SPAA), pp. 1–9 (1989). https://doi.org/10.1145/72935.72936
Dally, W.: Virtual-channel flow control. In: Proceedings of the 17th Annual International Symposium on Computer Architecture (ISCA), pp. 60–68 (1990). https://doi.org/10.1145/325164.325115
IEEE 802.11Qbb. Priority based flow control (2011)
Alizadeh, M., Greenberg, A., Maltz, D., et al.: Data Center TCP (DCTCP). In: ACM SIGCOMM (2010). https://doi.org/10.1145/1851275.1851192
Ramakrishnan, K., Floyd, S., Black, D.: The addition of explicit congestion notification (ECN). RFC 3168. https://doi.org/10.17487/RFC3168
Zhu, Y., Eran, H., Firestone, D., et al.: Congestion Control for Large-Scale RDMA Deployments. In: ACM SIGCOMM (2015). https://doi.org/10.1145/2829988.2787484
IEEE. 802.11Qau. Congestion notification (2010)
IBTA: InfiniBand Architecture Specification, Volume 1, Release 1.5. Available to IBTA members via. https://www.infinibandta.org
Gusat, M., Craddock, D., Denzel, W., et al.: Congestion control in infiniband networks. In: Hot Interconnects, pp. 158–159 (2005). https://doi.org/10.1109/CONECT.2005.14
Gran, E., Eimot, M., Reinemo, S.-A., et al.: First experiences with congestion control in InfiniBand hardware. In: International Parallel and Distributed Processing Symposium. (2010). https://doi.org/10.1109/IPDPS.2010.5470419
Mittal, R., Lam, V., Dukkipati, N., et al.: TIMELY: RTT-based congestion control for the datacenter. In: ACM SIGCOMM (2015). https://doi.org/10.1145/2785956.2787510
Kumar, G., Dukkipati, N., Jang, K., et al.: Swift: delay is simple and effective for congestion control in the datacenter. In: SIGCOMM 2020: Proceedings ACM Special Interest Group on Data Communication, pp. 514–528 (2020). https://doi.org/10.1145/3387514.3406591
Wang, Y., Lan, M., Zhao, T., et al.: Combining RTT and ECN for RoCEv2 protocol. In: HPCCT and BDAI 2020: Proceedings 2020 4th High Performance Computing and Cluster Technologies Conference and 2020 3rd International Conference on Big Data and Artificial Intelligence, pp. 158–164, Qingdao, China (2020). https://doi.org/10.1145/3409501.3409509
Li, Y., Miao, R., Liu, H., et al.: HPCC: high precision congestion control. In: SIGCOMM 2019: Proc. ACM Special Interest Group on Data Communication, pp. 44–58 (2019). https://doi.org/10.1145/3341302.3342085
Xue, J., Chaudhry, M., Vamanan, B., et al.: Dart: divide and specialize for fast response to congestion in RDMA-based datacenter networks. IEEE/ACM Trans. Networking 28(1), 322–335 (2020). https://doi.org/10.1109/TNET.2019.2961671
Yang, C., Reddy, A.: A taxonomy for congestion control algorithms in packet switching networks. IEEE Network 9(4), 34–45 (1995). https://doi.org/10.1109/65.397042
Saylor, D.: Evo: a hybrid optimizer employing evolutionary algorithms and reinforcement meta learning agents. [Unpublished manuscript]. Applied Machine Learning and Artificial Intelligence, NVIDIA (2013)
Effective Bandwidth Benchmark Homepage. https://fs.hlrs.de/projects/par/mpi/b_eff/b_eff_3.1
Incompact3D Homepage. https://www.incompact3d.com
Bartholomew, P., Deskos, G., et al.: Xcompact3D: an open-source framework for solving turbulence problems on a Cartesian mesh. SoftwareX 12, 100550 (2020). https://doi.org/10.1016/j.softx.2020.100550
LAMMPS Homepage. https://www.lammps.org
Thompson, A., Aktulga, H., et al.: LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales. Comp. Phys. Comm. 271, 10817, 100550 (2022). https://doi.org/10.1016/j.cpc.2021.108171
VASP Homepage. https://www.vasp.at
Kresse, G., Hafner, J.: Ab initio molecular dynamics for liquid metals. Phys. Rev. B 47, 558 (1993). https://doi.org/10.1016/0022-3093(95)00355-X
Kresse, G., Hafner, J.: Ab initio molecular-dynamics simulation of the liquid-metal-amorphous-semiconductor transition in germanium. Phys. Rev. B 49, 14251 (1994). https://doi.org/10.1103/PhysRevB.49.14251
Kresse, G., Furthmüller, J.: Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mat. Sci. 6, 15 (1996). https://doi.org/10.1016/0927-0256(96)00008-0
Kresse, G., Furthmüller, J.: Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169 (1996). https://doi.org/10.1103/PhysRevB.54.11169
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Switzerland AG
About this paper
Cite this paper
Shpigelman, Y., Shainer, G., Graham, R., Qin, Y., Cisneros-Stoianowski, G., Stunkel, C. (2022). NVIDIA’s Quantum InfiniBand Network Congestion Control Technology and Its Impact on Application Performance. In: Varbanescu, AL., Bhatele, A., Luszczek, P., Marc, B. (eds) High Performance Computing. ISC High Performance 2022. Lecture Notes in Computer Science, vol 13289. Springer, Cham. https://doi.org/10.1007/978-3-031-07312-0_2
Download citation
DOI: https://doi.org/10.1007/978-3-031-07312-0_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-07311-3
Online ISBN: 978-3-031-07312-0
eBook Packages: Computer ScienceComputer Science (R0)