ABSTRACT
Distributed synchronized GPU training is commonly used for deep learning. The resource constraint of using a fixed number of GPUs makes large-scale training jobs suffer from long queuing time for resource allocation, and lowers the cluster utilization. Adapting to resource elasticity can alleviate this but often introduces inconsistent model accuracy, due to lacking of capability to decouple model training procedure from resource allocation. We propose EasyScale, an elastic training system that achieves consistent model accuracy under resource elasticity for both homogeneous and heterogeneous GPUs. EasyScale preserves the data-parallel training behaviors strictly, traces the consistency-relevant factors carefully, utilizes the deep learning characteristics for EasyScaleThread abstraction and fast context-switching. To utilize heterogeneous cluster, EasyScale dynamically assigns workers based on the intra-/inter-job schedulers, minimizing load imbalance and maximizing aggregated job throughput. Deployed in an online serving cluster, EasyScale powers the training jobs to utilize idle GPUs opportunistically, improving overall cluster utilization by 62.1%.
- 2023. Auto kernel selection of NVIDIA cuDNN. https://docs.nvidia.com/deeplearning/cudnn/api/index.html#cudnnGetConvolutionBackwardDataAlgorithm_v7.Google Scholar
- 2023. Auto selection of cuBLAS in Ampere GPUs. https://docs.nvidia.com/cuda/cublas/index.html#cublas-GemmEx.Google Scholar
- 2023. Elastic Horovod. https://github.com/horovod/horovod/blob/master/docs/elastic.rst.Google Scholar
- 2023. ElasticDL, https://github.com/sql-machine-learning/elasticdl/.Google Scholar
- 2023. NCCL deterministic. https://github.com/NVIDIA/nccl/issues/157.Google Scholar
- 2023. NVIDIA Framework-determinism, https://github.com/NVIDIA/framework-determinism.Google Scholar
- 2023. PyTorch Reproducibility. https://pytorch.org/docs/stable/notes/randomness.html.Google Scholar
- 2023. torch.backends.cudnn.benchmark. https://pytorch.org/docs/stable/backends.html.Google Scholar
- 2023. TorchElastic. https://pytorch.org/docs/stable/distributed.elastic.html.Google Scholar
- Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, Manjunath Kudlur, Josh Levenberg, Rajat Monga, Sherry Moore, Derek G. Murray, Benoit Steiner, Paul Tucker, Vijay Vasudevan, Pete Warden, Martin Wicke, Yuan Yu, and Xiaoqiang Zheng. 2016. TensorFlow: A System for Large-Scale Machine Learning. In 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16). USENIX Association, Savannah, GA, 265--283.Google ScholarDigital Library
- Sanjith Athlur, Nitika Saran, Muthian Sivathanu, Ramachandran Ramjee, and Nipun Kwatra. 2022. Varuna: Scalable, Low-Cost Training of Massive Deep Learning Models. In Proceedings of the Seventeenth European Conference on Computer Systems (Rennes, France) (EuroSys '22). Association for Computing Machinery, New York, NY, USA, 472--487. Google ScholarDigital Library
- Zhihao Bai, Zhen Zhang, Yibo Zhu, and Xin Jin. 2020. PipeSwitch: Fast Pipelined Context Switching for Deep Learning Applications. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 499--514.Google Scholar
- Paul Barham, Aakanksha Chowdhery, Jeff Dean, Sanjay Ghemawat, Steven Hand, Daniel Hurt, Michael Isard, Hyeontaek Lim, Ruoming Pang, Sudip Roy, Brennan Saeta, Parker Schuh, Ryan Sepassi, Laurent Shafey, Chandu Thekkath, and Yonghui Wu. 2022. Pathways: Asynchronous Distributed Dataflow for ML. In Proceedings of Machine Learning and Systems, D. Marculescu, Y. Chi, and C. Wu (Eds.), Vol. 4. 430--449.Google Scholar
- Zhengda Bian, Shenggui Li, Wei Wang, and Yang You. 2021. Online Evolutionary Batch Size Orchestration for Scheduling Deep Learning Workloads in GPU Clusters (SC '21). Association for Computing Machinery, New York, NY, USA, Article 100, 15 pages. Google ScholarDigital Library
- Brendan Burns, Brian Grant, David Oppenheimer, Eric Brewer, and John Wilkes. 2016. Borg, Omega, and Kubernetes. ACM Queue 14 (2016), 70--93.Google ScholarDigital Library
- Paris Carbone, Asterios Katsifodimos, Stephan Ewen, Volker Markl, Seif Haridi, and Kostas Tzoumas. 2015. Apache flink: Stream and batch processing in a single engine. Bulletin of the IEEE Computer Society Technical Committee on Data Engineering 36, 4 (2015).Google Scholar
- Kevin Clark, Minh-Thang Luong, Quoc V. Le, and Christopher D. Manning. 2020. ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators. In 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26--30, 2020. OpenReview.net. https://openreview.net/forum?id=r1xMH1BtvBGoogle Scholar
- Jeffrey Dean and Sanjay Ghemawat. 2004. MapReduce: Simplified Data Processing on Large Clusters. In 6th Symposium on Operating Systems Design & Implementation (OSDI 04). USENIX Association, San Francisco, CA.Google Scholar
- Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. 2009. Imagenet: A large-scale hierarchical image database. In Computer Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on. IEEE, Miami, FL, USA, 248--255.Google ScholarCross Ref
- Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL-HLT 2019, Minneapolis, MN, USA, June 2--7, 2019, Volume 1 (Long and Short Papers). Association for Computational Linguistics, Minneapolis, Minnesota, USA, 4171--4186.Google Scholar
- Mark Everingham, Luc Van Gool, Christopher KI Williams, John Winn, and Andrew Zisserman. 2010. The pascal visual object classes (voc) challenge. International journal of computer vision 88, 2 (2010), 303--338.Google ScholarDigital Library
- Shiqing Fan, Yi Rong, Chen Meng, Zongyan Cao, Siyu Wang, Zhen Zheng, Chuan Wu, Guoping Long, Jun Yang, Lixue Xia, Lansong Diao, Xiaoyong Liu, and Wei Lin. 2021. DAPPLE: a pipelined data parallel approach for training large models. In PPoPP '21: 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, Virtual Event, Republic of Korea, February 27- March 3, 2021, Jaejin Lee and Erez Petrank (Eds.). ACM, 431--445. Google ScholarDigital Library
- Simos Gerasimou, Hasan Ferit Eniser, Alper Sen, and Alper Cakan. 2020. Importance-driven deep learning system testing. In 2020 IEEE/ACM 42nd International Conference on Software Engineering (ICSE). IEEE, Seoul, Korea (South), 702--713.Google Scholar
- Priya Goyal, Piotr Dollár, Ross B. Girshick, Pieter Noordhuis, Lukasz Wesolowski, Aapo Kyrola, Andrew Tulloch, Yangqing Jia, and Kaiming He. 2017. Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour. CoRR abs/1706.02677 (2017). arXiv:1706.02677 http://arxiv.org/abs/1706.02677Google Scholar
- Juncheng Gu, Mosharaf Chowdhury, Kang G. Shin, and Aditya Akella. 2022. Elastic Model Aggregation with Parameter Service. CoRR abs/2204.03211 (2022). arXiv:2204.03211 https://arxiv.org/abs/2204.03211Google Scholar
- Mingcong Han, Hanze Zhang, Rong Chen, and Haibo Chen. 2022. Microsecond-scale Preemption for Concurrent GPU-accelerated DNN Inferences. In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22). USENIX Association, Carlsbad, CA, 539--558. https://www.usenix.org/conference/osdi22/presentation/hanGoogle Scholar
- F Maxwell Harper and Joseph A Konstan. 2015. The movielens datasets: History and context. Acm transactions on interactive intelligent systems (tiis) 5, 4 (2015), 1--19.Google Scholar
- K. He, X. Zhang, S. Ren, and J. Sun. 2016. Deep Residual Learning for Image Recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Computer Society, Los Alamitos, CA, USA, 770--778. Google ScholarCross Ref
- Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 2017. Neural Collaborative Filtering. In Proceedings of the 26th International Conference on World Wide Web (Perth, Australia) (WWW '17). International World Wide Web Conferences Steering Committee, Republic and Canton of Geneva, CHE, 173--182. Google ScholarDigital Library
- Michael Isard, Mihai Budiu, Yuan Yu, Andrew Birrell, and Dennis Fetterly. 2007. Dryad: Distributed Data-Parallel Programs from Sequential Building Blocks. In Proceedings of the 2nd ACM SIGOPS/EuroSys European Conference on Computer Systems 2007 (Lisbon, Portugal) (EuroSys '07). Association for Computing Machinery, New York, NY, USA, 59--72. Google ScholarDigital Library
- Myeongjae Jeon, Shivaram Venkataraman, Amar Phanishayee, Junjie Qian, Wencong Xiao, and Fan Yang. 2019. Analysis of Large-Scale Multi-Tenant GPU Clusters for DNN Training Workloads. In 2019 USENIX Annual Technical Conference (USENIX ATC 19). 947--960.Google Scholar
- Myeongjae Jeon, Shivaram Venkataraman, Junjie Qian, Amar Phanishayee, Wencong Xiao, and Fan Yang. 2018. Multi-tenant GPU clusters for deep learning workloads: Analysis and implications. Tech. Rep. (2018).Google Scholar
- Xianyan Jia, Le Jiang, Ang Wang, Wencong Xiao, Ziji Shi, Jie Zhang, Xinyuan Li, Langshi Chen, Yong Li, Zhen Zheng, Xiaoyong Liu, and Wei Lin. 2022. Whale: Efficient Giant Model Training over Heterogeneous GPUs. In 2022 USENIX Annual Technical Conference (USENIX ATC 22). USENIX Association, Carlsbad, CA, 673--688.Google Scholar
- Shen Li, Yanli Zhao, Rohan Varma, Omkar Salpekar, Pieter Noordhuis, Teng Li, Adam Paszke, Jeff Smith, Brian Vaughan, Pritam Damania, and Soumith Chintala. 2020. PyTorch Distributed: Experiences on Accelerating Data Parallel Training. Proc. VLDB Endow. 13, 12 (aug 2020), 3005--3018. Google ScholarDigital Library
- Gangmuk Lim, Jeongseob Ahn, Wencong Xiao, Youngjin Kwon, and Myeongjae Jeon. 2021. Zico: Efficient GPU Memory Sharing for Concurrent DNN Training. In 2021 USENIX Annual Technical Conference (USENIX ATC 21). USENIX Association, 161--175.Google Scholar
- Wei Lin, Zhengping Qian, Junwei Xu, Sen Yang, Jingren Zhou, and Lidong Zhou. 2016. StreamScope: Continuous Reliable Distributed Processing of Big Data Streams. In 13th USENIX Symposium on Networked Systems Design and Implementation (NSDI 16). USENIX Association, Santa Clara, CA, 439--453.Google ScholarDigital Library
- Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. 2021. Swin Transformer: Hierarchical Vision Transformer using Shifted Windows. CoRR abs/2103.14030 (2021). arXiv:2103.14030 https://arxiv.org/abs/2103.14030Google Scholar
- Lingxiao Ma, Zhiqiang Xie, Zhi Yang, Jilong Xue, Youshan Miao, Wei Cui, Wenxiang Hu, Fan Yang, Lintao Zhang, and Lidong Zhou. 2020. Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 881--897. https://www.usenix.org/conference/osdi20/presentation/maGoogle Scholar
- Ningning Ma, Xiangyu Zhang, Hai-Tao Zheng, and Jian Sun. 2018. Shufflenet v2: Practical guidelines for efficient cnn architecture design. In Proceedings of the European conference on computer vision (ECCV). IEEE, Munich, Germany, 116--131.Google ScholarDigital Library
- Luo Mai, Guo Li, Marcel Wagenländer, Konstantinos Fertakis, Andrei-Octavian Brabete, and Peter Pietzuch. 2020. KungFu: Making Training in Distributed Machine Learning Adaptive. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 937--954.Google Scholar
- Prabhat Nagarajan, Garrett Warnell, and Peter Stone. 2018. Deterministic implementations for reproducibility in deep reinforcement learning. arXiv preprint arXiv:1809.05676 (2018).Google Scholar
- Deepak Narayanan, Aaron Harlap, Amar Phanishayee, Vivek Seshadri, Nikhil R. Devanur, Gregory R. Ganger, Phillip B. Gibbons, and Matei Zaharia. 2019. PipeDream: Generalized Pipeline Parallelism for DNN Training. In Proceedings of the 27th ACM Symposium on Operating Systems Principles (Huntsville, Ontario, Canada) (SOSP '19). Association for Computing Machinery, New York, NY, USA, 1--15. Google ScholarDigital Library
- Deepak Narayanan, Keshav Santhanam, Fiodar Kazhamiaka, Amar Phanishayee, and Matei Zaharia. 2020. Heterogeneity-Aware Cluster Scheduling Policies for Deep Learning Workloads. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 481--498.Google Scholar
- Andrew Or, Haoyu Zhang, and Michael None Freedman. 2022. VirtualFlow: Decoupling Deep Learning Models from the Underlying Hardware. In Proceedings of Machine Learning and Systems, D. Marculescu, Y. Chi, and C. Wu (Eds.), Vol. 4. 126--140.Google Scholar
- Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al. 2019. PyTorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems. 8024--8035.Google Scholar
- Yanghua Peng, Yixin Bao, Yangrui Chen, Chuan Wu, and Chuanxiong Guo. 2018. Optimus: An Efficient Dynamic Resource Scheduler for Deep Learning Clusters. In Proceedings of the Thirteenth European Conference on Computer Systems. ACM, New York, NY, USA.Google ScholarDigital Library
- Hung Viet Pham, Shangshu Qian, Jiannan Wang, Thibaud Lutellier, Jonathan Rosenthal, Lin Tan, Yaoliang Yu, and Nachiappan Nagappan. 2021. Problems and Opportunities in Training Deep Learning Software Systems: An Analysis of Variance. In Proceedings of the 35th IEEE/ACM International Conference on Automated Software Engineering (Virtual Event, Australia) (ASE '20). Association for Computing Machinery, New York, NY, USA, 771--783. Google ScholarDigital Library
- Shangshu Qian, Hung Pham, Thibaud Lutellier, Zeou Hu, Jungwon Kim, Lin Tan, Yaoliang Yu, Jiahao Chen, and Sameena Shah. 2021. Are My Deep Learning Systems Fair? An Empirical Study of Fixed-Seed Training. Advances in Neural Information Processing Systems 34 (2021).Google Scholar
- Aurick Qiao, Sang Keun Choe, Suhas Jayaram Subramanya, Willie Neiswanger, Qirong Ho, Hao Zhang, Gregory R. Ganger, and Eric P. Xing. 2021. Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning. In 15th USENIX Symposium on Operating Systems Design and Implementation (OSDI 21). USENIX Association, 1--18.Google Scholar
- Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, and Yuxiong He. 2020. Zero: Memory optimizations toward training trillion parameter models. In SC20: International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE, Atlanta, GA, USA, 1--16.Google ScholarCross Ref
- Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. SQuAD: 100, 000+ Questions for Machine Comprehension of Text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, EMNLP 2016, Austin, Texas, USA, November 1--4, 2016. The Association for Computational Linguistics, 2383--2392. Google ScholarCross Ref
- Joseph Redmon and Ali Farhadi. 2018. YOLOv3: An Incremental Improvement. CoRR abs/1804.02767 (2018). arXiv:1804.02767 http://arxiv.org/abs/1804.02767Google Scholar
- Dharma Shukla, Muthian Sivathanu, Srinidhi Viswanatha, Bhargav Gulavani, Rimma Nehme, Amey Agrawal, Chen Chen, Nipun Kwatra, Ramachandran Ramjee, Pankaj Sharma, et al. 2022. Singularity: Planet-Scale, Preemptible, Elastic Scheduling of AI Workloads. CoRR abs/2202.07848 (2022).Google Scholar
- Karen Simonyan and Andrew Zisserman. 2015. Very Deep Convolutional Networks for Large-Scale Image Recognition. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7--9, 2015, Conference Track Proceedings. http://arxiv.org/abs/1409.1556Google Scholar
- Muthian Sivathanu, Tapan Chugh, Sanjay S. Singapuram, and Lidong Zhou. 2019. Astra: Exploiting Predictability to Optimize Deep Learning. In Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems (Providence, RI, USA) (ASPLOS '19). Association for Computing Machinery, New York, NY, USA, 909--923. Google ScholarDigital Library
- Abhishek Verma, Luis Pedrosa, Madhukar Korupolu, David Oppenheimer, Eric Tune, and John Wilkes. 2015. Large-scale cluster management at Google with Borg. In Proceedings of the Tenth European Conference on Computer Systems. ACM, New York, NY, USA, 18.Google ScholarDigital Library
- Qizhen Weng, Wencong Xiao, Yinghao Yu, Wei Wang, Cheng Wang, Jian He, Yong Li, Liping Zhang, Wei Lin, and Yu Ding. 2022. MLaaS in the Wild: Workload Analysis and Scheduling in Large-Scale Heterogeneous GPU Clusters. In 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI 22). USENIX Association, Renton, WA, 945--960.Google Scholar
- Yidi Wu, Kaihao Ma, Xiao Yan, Zhi Liu, Zhenkun Cai, Yuzhen Huang, James Cheng, Han Yuan, and Fan Yu. 2022. Elastic Deep Learning in Multi-Tenant GPU Clusters. IEEE Transactions on Parallel and Distributed Systems 33, 1 (2022), 144--158. Google ScholarCross Ref
- Wencong Xiao, Romil Bhardwaj, Ramachandran Ramjee, Muthian Sivathanu, Nipun Kwatra, Zhenhua Han, Pratyush Patel, Xuan Peng, Hanyu Zhao, Quanlu Zhang, Fan Yang, and Lidong Zhou. 2018. Gandiva: Introspective Cluster Scheduling for Deep Learning. In 13th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2018, Carlsbad, CA, USA, October 8--10, 2018. USENIX Association, 595--610. https://www.usenix.org/conference/osdi18/presentation/xiaoGoogle Scholar
- Wencong Xiao, Shiru Ren, Yong Li, Yang Zhang, Pengyang Hou, Zhi Li, Yihui Feng, Wei Lin, and Yangqing Jia. 2020. AntMan: Dynamic Scaling on GPU Clusters for Deep Learning. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 533--548.Google Scholar
- Matei Zaharia, Mosharaf Chowdhury, Tathagata Das, Ankur Dave, Justin Ma, Murphy McCauley, Michael J. Franklin, Scott Shenker, and Ion Stoica. 2012. Resilient Distributed Datasets: A Fault-tolerant Abstraction for In-memory Cluster Computing. In Proceedings of the 9th USENIX Conference on Networked Systems Design and Implementation (San Jose, CA) (NSDI'12). USENIX, 1 pages.Google ScholarDigital Library
- Haoyu Zhang, Logan Stafman, Andrew Or, and Michael J Freedman. 2017. SLAQ: quality-driven scheduling for distributed machine learning. In Proceedings of the 2017 Symposium on Cloud Computing. ACM, New York, NY, USA, 390--404.Google ScholarDigital Library
- Quanlu Zhang, Zhenhua Han, Fan Yang, Yuge Zhang, Zhe Liu, Mao Yang, and Lidong Zhou. 2020. Retiarii: A Deep Learning Exploratory-Training Framework. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 919--936.Google Scholar
- Shuai Zhang, Lina Yao, Aixin Sun, and Yi Tay. 2019. Deep learning based recommender system: A survey and new perspectives. ACM Computing Surveys (CSUR) 52, 1 (2019), 1--38.Google ScholarDigital Library
- Hanyu Zhao, Zhenhua Han, Zhi Yang, Quanlu Zhang, Fan Yang, Lidong Zhou, Mao Yang, Francis C.M. Lau, Yuqi Wang, Yifan Xiong, and Bin Wang. 2020. HiveD: Sharing a GPU Cluster for Deep Learning with Guarantees. In 14th USENIX Symposium on Operating Systems Design and Implementation (OSDI 20). USENIX Association, 515--532.Google Scholar
Index Terms
- EasyScale: Elastic Training with Consistent Accuracy and Improved Utilization on GPUs
Recommendations
Optimizing makespan and resource utilization for multi-DNN training in GPU cluster
AbstractDeep neural network (DNN) has been widely applied in many fields of artificial intelligence (AI), gaining great popularity both in industry and academia. Increasing the size of DNN models does dramatically improve the learning ...
Highlights- OMRU is proposed to optimize makespan and resource utilization for multi-DNN training in GPU cluster.
Dynamic Resource Management for Efficient Utilization of Multitasking GPUs
ASPLOS '17As graphics processing units (GPUs) are broadly adopted, running multiple applications on a GPU at the same time is beginning to attract wide attention. Recent proposals on multitasking GPUs have focused on either spatial multitasking, which partitions ...
Comments