Abstract
Scaling deep learning workloads across multiple GPUs on a single node has become increasingly important in data analytics. A key question is how well a PCIe-based GPU interconnect can perform relative to a custom high-performance interconnect such as NVIDIA’s NVLink. This paper evaluates two such on-node interconnects for eight NVIDIA Pascal P100 GPUs: (a) the NVIDIA DGX-1’s NVLink 1.0 ‘hybrid cube mesh’; and (b) the Cirrascale GX8’s two-level PCIe tree using dual SR3615 switch risers. To show the effects of a range of neural network workloads, we define a parameterized version of the popular ResNet architecture. We define a workload intensity metric that characterizes the expected computation/communication ratio; we also locate AlexNet and GoogLeNet within that space. As expected, the DGX-1 typically has superior performance. However, the GX8 is very competitive on all ResNet workloads. With 8 GPUs, the GX8 can outperform the DGX-1 on all-to-all reductions by 10% for medium-sized payloads; and in rare cases, the GX8 slightly outperforms on ResNet.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Gawande, N.A., Landwehr, J.B., Daily, J.A., Tallent, N.R., Vishnu, A., Kerbyson, D.J.: Scaling deep learning workloads: NVIDIA DGX-1/Pascal and Intel Knights Landing. In: 2017 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pp. 399–408, May 2017
Foley, D., Danskin, J.: Ultra-performance Pascal GPU and NVLink interconnect. IEEE Micro 37(2), 7–17 (2017)
Sodani, A., Gramunt, R., Corbal, J., Kim, H.S., Vinod, K., Chinthamani, S., Hutsell, S., Agarwal, R., Liu, Y.C.: Knights landing: second-generation Intel Xeon Phi product. IEEE Micro 36(2), 34–46 (2016)
Cirrascale: The GX8 series multi-device peering platform, June 2016. http://www.cirrascale.com/documents/datasheets/Cirrascale_GX8Series_Datasheet_CM080C.pdf
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Pereira, F., Burges, C., Bottou, L., Weinberger, K. (eds.) Advances in Neural Information Processing Systems, vol. 25, pp. 1097–1105. Curran Associates, Inc., Red Hook (2012)
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., Rabinovich, A.: Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9 (2015)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)
Cirrascale: Cirrascale SR3615 PCIe switch riser, July 2015
Cirrascale: Scaling GPU compute performance (2015). http://www.cirrascale.com/documents/whitepapers/Cirrascale_ScalingGPUCompute_WP_M987_REVA.pdf
Luehr, N.: Fast multi-GPU collectives with NCCL, April 2016. https://devblogs.nvidia.com/parallelforall/fast-multi-gpu-collectives-nccl/
NVIDIA: NCCL: NVIDIA collective communications library, August 2017. https://developer.nvidia.com/nccl
Awan, A.A., Hamidouche, K., Venkatesh, A., Panda, D.K.: Efficient large message broadcast using NCCL and CUDA-aware MPI for deep learning. In: Proceedings of the 23rd European MPI Users’ Group Meeting, EuroMPI 2016, New York, NY, USA, pp. 15–22. ACM (2016)
Jeaugey, S.: Optimized inter-GPU collective operations with NCCL, May 2017. http://on-demand-gtc.gputechconf.com/gtc-quicklink/8Bdyh
Patarasuk, P., Yuan, X.: Bandwidth optimal all-reduce algorithms for clusters of workstations. J. Parallel Distrib. Comput. 69(2), 117–124 (2009)
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., Berg, A.C., Fei-Fei, L.: Imagenet large scale visual recognition challenge. Int. J. Comput. Vis. 115(3), 211–252 (2015)
Berkeley Vision and Learning Center. Berkeley vision and learning center: Caffe (2016). http://caffe.berkeleyvision.org
Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., Darrell, T.: Caffe: Convolutional architecture for fast feature embedding. arXiv preprint arXiv:1408.5093 (2014)
Berkeley Vision and Learning Center: Convolutional architecture for fast feature embedding (Caffe) (2016). https://github.com/BVLC/caffe/
NVIDIA: Convolutional architecture for fast feature embedding (Caffe) (2017). https://github.com/NVIDIA/caffe
Ben-Nun, T.: MGBench: multi-GPU computing benchmark suite, February 2016. https://github.com/tbennun/mgbench
Cirrascale: Cirrascale SR3514: Unexpected performance inequality. Technical Brief M901A–092014
Adolf, R., Rama, S., Reagen, B., Wei, G.Y., Brooks, D.: Fathom: reference workloads for modern deep learning methods. In: 2016 IEEE International Symposium on Workload Characterization (IISWC), pp. 1–10, September 2016
Christensen, C., Fogal, T., Luehr, N., Woolley, C.: Topology-aware image compositing using NVLink. In: 2016 IEEE 6th Symposium on Large Data Analysis and Visualization (LDAV), pp. 93–94, October 2016
Shams, S., Platania, R., Lee, K., Park, S.J.: Evaluation of deep learning frameworks over different HPC architectures. In: Proceedings of the IEEE 37th International Conference on Distributed Computing Systems, pp. 1389–1396, June 2017
Nomura, S., Mitsuishi, T., Suzuki, J., Hayashi, Y., Kan, M., Amano, H.: Performance analysis of the multi-GPU system with expether. SIGARCH Comput. Archit. News 42(4), 9–14 (2014)
Ben-Nun, T., Sutton, M., Pai, S., Pingali, K.: Groute: an asynchronous multi-GPU programming model for irregular computations. In: Proceedings of the 22nd ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, New York, NY, USA, pp. 235–248. ACM (2017)
Awan, A.A., Chu, C.H., Subramoni, H., Panda, D.K.: Optimized broadcast for deep learning workloads on dense-GPU infiniband clusters: MPI or NCCL? arXiv preprint arXiv:1707.09414 (2017)
Acknowledgments
The authors thank Matthew Macduff (PNNL) for evaluation assistance. We are grateful for funding support from the U.S. Department of Energy’s (DOE) Office of Advanced Scientific Computing Research as part of the “Center for Advanced Technology Evaluation” (CENATE) and “Convergence of Deep Learning and Machine Learning for HPC Simulation and Modeling.” Pacific Northwest National Laboratory is operated by Battelle for the DOE under Contract DE-AC05-76RL01830.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this paper
Cite this paper
Tallent, N.R., Gawande, N.A., Siegel, C., Vishnu, A., Hoisie, A. (2018). Evaluating On-Node GPU Interconnects for Deep Learning Workloads. In: Jarvis, S., Wright, S., Hammond, S. (eds) High Performance Computing Systems. Performance Modeling, Benchmarking, and Simulation. PMBS 2017. Lecture Notes in Computer Science(), vol 10724. Springer, Cham. https://doi.org/10.1007/978-3-319-72971-8_1
Download citation
DOI: https://doi.org/10.1007/978-3-319-72971-8_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-72970-1
Online ISBN: 978-3-319-72971-8
eBook Packages: Computer ScienceComputer Science (R0)