FairGV: Fair and fast GPU virtualization

IEEE Transactions on Parallel and Distributed Systems

Volumn 28, Issue 12, 2017, Pages 3472-3485

  Hong, Cheol Ho ^a   Spence, Ivor ^a   Nikolopoulos, Dimitrios S ^a  

^a QUEEN'S UNIVERSITY BELFAST   (United Kingdom)

Author keywords

coscheduling and hybrid scheduling strategies; fair queuing; GPU virtualization; trap less architecture

Indexed keywords

COMPUTER ARCHITECTURE; PROGRAM PROCESSORS; SCHEDULING; VIRTUAL REALITY; VIRTUALIZATION;

COLLABORATIVE SCHEDULING; FAIR QUEUING; HIGH PERFORMANCE COMPUTING (HPC); HYBRID SCHEDULING STRATEGIES; PERFORMANCE DEGRADATION; PERFORMANCE LIMITATIONS; PROTOTYPE IMPLEMENTATIONS; VIRTUALIZATION SOFTWARE;

GRAPHICS PROCESSING UNIT;

EID: 85021813719     PISSN: 10459219     EISSN: None     Source Type: Journal    
DOI: 10.1109/TPDS.2017.2717908     Document Type: Article

References (39)

1. N. Satish, et al., "Fast sort on CPUs and GPUs: A case for bandwidth oblivious SIMD sort, " in Proc. ACM SIGMOD Int. Conf. Manage. Data, 2010, pp. 351-362.
2. M. Silberstein, B. Ford, I. Keidar, and E. Witchel, "GPUfs: Integrating a file system with GPUs, " ACM SIGARCH Comput. Archit. News, vol. 41, no. 1, pp. 485-498, 2013.
3. Q. He, S. Zhou, B. Kobler, D. Duffy, and T. McGlynn, "Case study for running HPC applications in public clouds, " in Proc. 19th ACM Int. Symp. High Perform. Distrib. Comput., 2010, pp. 395-401.
4. NVIDIA, "GP100 pascal whitepaper, " 2016. [Online]. Available: https://images. nvidia. com/content/pdf/tesla/whitepaper/ pascal-architecture-whitepaper. pdf
5. I. Tanasic, I. Gelado, J. Cabezas, A. Ramirez, N. Navarro, and M. Valero, "Enabling preemptive multiprogramming on GPUs, " ACM SIGARCH Comput. Archit. News, vol. 42, no. 3, pp. 193-204, 2014.
6. V. Gupta, et al., "GViM: GPU-accelerated virtual machines, " in Proc. 3rd ACM Workshop System-Level Virtualization High Perform. Comput., 2009, pp. 17-24.
7. G. Giunta, R. Montella, G. Agrillo, and G. Coviello, "A GPGPU transparent virtualization component for high performance computing clouds, " in Euro-Par 2010-Parallel Processing. Berlin, Germany: Springer, 2010, pp. 379-391.
8. S. Kato, K. Lakshmanan, R. Rajkumar, and Y. Ishikawa, "TimeGraph: GPU scheduling for real-time multi-tasking environments, " in Proc. USENIX Annu. Tech. Conf., 2011, Art. no. 17.
9. S. Kato, M. McThrow, C. Maltzahn, and S. A. Brandt, "Gdev: Firstclass GPU resource management in the operating system, " in Proc. USENIX Annu. Tech. Conf., 2012, pp. 401-412.
10. V. Gupta, K. Schwan, N. Tolia, V. Talwar, and P. Ranganathan, "Pegasus: Coordinated scheduling for virtualized acceleratorbased systems, " in Proc. USENIX Annu. Tech. Conf., 2011, Art. no. 31.
11. L. Shi, H. Chen, J. Sun, and K. Li, "vCUDA: GPU-accelerated highperformance computing in virtual machines, " IEEE Trans. Comput., vol. 61, no. 6, pp. 804-816, Jun. 2012.
12. Y. Suzuki, S. Kato, H. Yamada, and K. Kono, "GPUvm: Why not virtualizing GPUs at the hypervisor?" in Proc. USENIX Annu. Tech. Conf., 2014, pp. 109-120.
13. J. Duato, A. J. Pena, F. Silla, R. Mayo, and E. S. Quintana-Ort -?, "rCUDA: Reducing the number of GPU-based accelerators in high performance clusters, " in Proc. Int. Conf. High Perform. Comput. Simul., 2010, pp. 224-231.
14. K. Menychtas, K. Shen, and M. L. Scott, "Disengaged scheduling for fair, protected access to fast computational accelerators, " ACM SIGPLAN Notices, vol. 49, no. 4, pp. 301-316, 2014.
15. C.-H. Hong, I. Spence, and S. D. Nikolopoulos, "GPU virtualization and scheduling methods: A comprehensive survey, " ACM Comput. Surveys (CSUR), vol. 50, no. 3, p. 35, 2017.
16. C. Lee, S.-W. Kim, and C. Yoo, "VADI: GPU virtualization for an automotive platform, " IEEE Trans. Ind. Informat., vol. 12, no. 1, pp. 277-290, Feb. 2016.
17. J. Duato, A. J. Pena, F. Silla, J. C. Fernandez, R. Mayo, and E. S. Quintana-Ort -?, "Enabling CUDA acceleration within virtual machines using rCUDA, " in Proc. 18th Int. Conf. High Perform. Comput., 2011, pp. 1-10.
18. R. Montella, et al., "On the virtualization of CUDA based GPU remoting on ARM and X86 machines in the GVirtuS framework, " Int. J. Parallel Program., pp. 1-22, 2016. [Online]. Available: http:// dx. doi. org/10. 1007/s10766-016-0462-1
19. X. Foundation, "Nouveau: Accelerated open source driver for NVIDIA cards, " 2011. [Online]. Available: https://nouveau. freedesktop. org/wiki/
20. PathScale, "pathscale/pscnv, " 2012. [Online]. Available: https:// github. com/pathscale/pscnv
21. K. Tian, Y. Dong, and D. Cowperthwaite, "A full GPU virtualization solution with mediated pass-through, " in Proc. USENIX Conf. USENIX Annu. Tech. Conf., 2014, pp. 121-132.
22. J. Song, Z. Lv, and K. Tian, "KVMGT: A full GPU virtualization solution, " in KVM Forum, 2014, http://www. linux-kvm. org/ page/KVM-Forum-2014
23. A. Herrera, "NVIDIA grid: Graphics accelerated VDI with the visual performance of a workstation, " NVIDIA Corp, Santa Clara, CA, USA, 2014.
24. M. Dowty and J. Sugerman, "GPU virtualization on VMware's hosted I/O architecture, " ACM SIGOPS Operating Syst. Rev., vol. 43, no. 3, pp. 73-82, 2009.
25. L. Soares and M. Stumm, "FlexSC: Flexible system call scheduling with exception-less system calls, " in Proc. 9th USENIX Conf. Operating Syst. Des. Implementation, 2010, pp. 1-8.
26. P. Barham, et al., "Xen and the art of virtualization, " ACM SIGOPS Operating Syst. Rev., vol. 37, no. 5, pp. 164-177, 2003.
27. J. Giacomoni, T. Moseley, and M. Vachharajani, "FastForward for efficient pipeline parallelism: A cache-optimized concurrent lockfree queue, " in Proc. 13th ACM SIGPLAN Symp. Principles Practice Parallel Program., 2008, pp. 43-52.
28. C. Li, C. Ding, and K. Shen, "Quantifying the cost of context switch, " in Proc. Workshop Exp. Comput. Sci., 2007, Art. no. 2.
29. N. Har'El, A. Gordon, A. Landau, M. Ben-Yehuda, A. Traeger, and R. Ladelsky, "Efficient and scalable paravirtual I/O system, " in Proc. USENIX Annu. Tech. Conf., 2013, pp. 231-242.
30. A. G. Greenberg and N. Madras, "How fair is fair queuing, " J. ACM, vol. 39, no. 3, pp. 568-598, 1992.
31. J. K. Ousterhout, "Scheduling techniques for concurrent systems, " in Proc. 3rd Int. Conf. Distrib. Comput. Syst., 1982, pp. 22-30.
32. C. Weng, Q. Liu, L. Yu, and M. Li, "Dynamic adaptive scheduling for virtual machines, " in Proc. 20th Int. Symp. High Perform. Distrib. Comput., 2011, pp. 239-250.
33. H. Kim, H. Lim, J. Jeong, H. Jo, and J. Lee, "Task-aware virtual machine scheduling for I/O performance, " in Proc. ACM SIGPLAN/SIGOPS Int. Conf. Virtual Execution Environments, 2009, pp. 101-110.
34. C. S. Pabla, "Completely fair scheduler, " Linux J., vol. 2009, no. 184, 2009, Art. no. 4.
35. K. Menychtas, K. Shen, and M. L. Scott, "Enabling OS research by inferring interactions in the black-box GPU stack, " in Proc. USENIX Annu. Tech. Conf., 2013, pp. 291-296.
36. C. Margiolas and M. F. O'Boyle, "Portable and transparent software managed scheduling on accelerators for fair resource sharing, " in Proc. Int. Symp. Code Generation Optimization, 2016, pp. 82-93.
37. H. Zhou, G. Tong, and C. Liu, "GPES: A preemptive execution system for GPGPU computing, " in Proc. 21st IEEE Real-Time Embedded Technol. Appl. Symp., 2015, pp. 87-97.
38. S. Che, et al., "Rodinia: A benchmark suite for heterogeneous computing, " in Proc. IEEE Int. Symp. Workload Characterization, 2009, pp. 44-54.
39. D. Shue, M. J. Freedman, and A. Shaikh, "Performance isolation and fairness for multi-tenant cloud storage, " in Proc. 10th USENIX Conf. Operating Syst. Des. Implementation, 2012, pp. 349-362.