KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Management Technique of Remote VR Service for Multiple Users in Container-Based Cloud Environments Sharing GPU

GPU를 공유하는 컨테이너 기반 클라우드 환경에서 다수의 사용자를 위한 원격 VR 서비스의 성능 관리 기법

  • 강지훈 (고려대학교 4단계 BK21 컴퓨터학교육연구단)
  • Received : 2021.10.26
  • Accepted : 2021.11.15
  • Published : 2022.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

Virtual Reality(VR) technology is an interface technology that is actively used in various audio-visual-based applications by showing users a virtual world composed of computer graphics. Since VR-based applications are graphic processing-based applications, expensive computing devices equipped with Graphics Processing Unit(GPU) are essential for graphic processing. This incurs a cost burden on VR application users for maintaining and managing computing devices, and as one of the solutions to this, a method of operating services in cloud environments is being used. This paper proposes a performance management technique to address the problem of performance interference between containers owing to GPU resource competition in container-based high-performance cloud environments in which multiple containers share a single GPU. The proposed technique reduces performance deviation due to performance interference, helping provide uniform performance-based remote VR services for users. In addition, this paper verifies the efficiency of the proposed technique through experiments.

VR(Virtual Reality) 기술은 사용자에게 컴퓨터 그래픽으로 구성된 가상 세계를 보여줌으로써 다양한 시청각 기반 응용에 적극적으로 활용되는 인터페이스 기술이다. VR 기반 응용은 그래픽 처리 기반 응용이기 때문에 그래픽 처리를 위해 GPU(Graphics Processing Unit)가 장착된 고가의 컴퓨팅 장치가 필수적으로 요구된다. 이는 VR 응용 사용자에게 컴퓨팅 장치의 유지, 관리에 대한 비용 부담을 발생시키며, 이를 해결하는 방법의 하나로써 서비스를 클라우드 환경에서 운용하는 방법이 사용되고 있다. 본 논문에서는 다수의 컨테이너가 VR 응용을 실행하기 위해 GPU를 공유하는 컨테이너 기반 고성능 클라우드 환경에서 GPU 자원 경쟁으로 인해 발생하는 컨테이너 사이의 성능 간섭 문제를 해결하기 위한 성능 관리 기법을 제안한다. 제안하는 기법은 성능 간섭으로 인한 성능 편차를 감소시켜 사용자에게 균일한 성능의 클라우드 기반 원격 VR 서비스를 제공할 수 있도록 지원한다. 또한, 본 논문에서는 실험을 통해 제안하는 기법의 효율성을 검증한다.

Keywords

References

  1. Oculus, Oculus Quest2 [Internet], https://www.oculus.com/quest-2/
  2. HTC Vive, HTC Vive Pro2 [Internet], https://www.vive.com/kr/product/vive-pro2-full-kit/overview/
  3. Amazon, Amazon EC2 Instance Types [Internet], https://aws.amazon.com/ec2/instance-types/?nc1=f_ls
  4. Microsoft, Azure NCv3 [Internet], https://docs.microsoft.com/ko-kr/azure/virtual-machines/ncv3-series
  5. Docker, Docker [Internet], https://www.docker.com/
  6. NVIDIA, NVIDIA Docker [Internet], https://github.com/NVIDIA/nvidia-docker
  7. S. Kato, K. Lakshmanan, R. Rajkumar, and Y. Ishikawa, "TimeGraph: GPU scheduling for real-time multi-tasking environments," USENIX Annual Technical Conference, 2011, pp.17-30.
  8. X. Long, X. Gong, Y. Liu, X. Que, and W. Wang, "Toward OS-level and device-level cooperative scheduling for multitasking GPUs," IEEE Access, pp.65711-65725, 2020. https://doi.org/10.1109/access.2020.2983731
  9. NVIDIA, NVIDIA GRID [Internet], https://www.nvidia.com/en-us/data-center/virtual-pc-apps/
  10. AMD, AMD Radeon Pro [Internet], https://www.amd.com/ko/products/server-accelerators/amd-radeon-pro-v520
  11. NVIDIA, NVIDIA Docker Wiki [Internet], https://github.com/NVIDIA/nvidia-docker/wiki/Frequently-Asked-Questions#i-have-multiple-gpu-devices-how-can-i-isolate-them-between-my-containers
  12. D. Abramson, et al., "Intel virtualization technology for directed I/O," Intel Technology Journal, Vol.10, No.3, pp.179-192, 2006.
  13. Y. Suzuki, S. Kato, H. Yamada, and K. Kono, "GPUvm: Why not virtualizing GPUs at the hypervisor?," USENIX Annual Technical Conference, pp.109-120, 2014.
  14. K. Tian, Y. Dong, and D. Cowperthwaite, "A full GPU virtualization solution with mediated pass-through," USENIX Annual Technical Conference, pp.121-132, 2014.
  15. H. Tan, Y. Tan, X. He, K. Li, and K. Li, "A virtual multi-channel GPU fair scheduling method for virtual machines," IEEE Transactions on Parallel and Distributed Systems, Vol.30, No.2, pp.257-270, 2018. https://doi.org/10.1109/TPDS.2018.2865341
  16. X. Zhao, J. Yao, P. Gao, and H. Guan, "Efficient sharing and fine-grained scheduling of virtualized GPU resources," IEEE 38th International Conference on Distributed Computing Systems (ICDCS), pp.742-752, 2018.
  17. Google, Google Stadia [Internet], https://stadia.google.com/
  18. NVIDIA, Geforce NOW [Internet], https://www.nvidia.com/ko-kr/geforce-now/
  19. T. Yoshihara and S. Fujita, "Fog-assisted virtual reality MMOG with ultra low latency," Seventh International Symposium on Computing and Networking (CANDAR), pp.121-129, 2019.
  20. M. Viitanen, J. Vanne, T. D. Hamalainen, and A. Kulmala, "Low latency edge rendering scheme for interactive 360 degree virtual reality gaming," IEEE 38th International Conference on Distributed Computing Systems (ICDCS), pp.1557-1560, 2018.
  21. A. Alshahrani, I. A. Elgendy, A. Muthanna, A. M. Alghamdi, and A. Alshamrani, "Efficient multi-player computation offloading for VR edge-cloud computing systems," Applied Sciences, Vol.10, No.16, pp.5515, 2020. https://doi.org/10.3390/app10165515
  22. H. Zhang, J. Zhang, X. Yin, K. Zhou, and Z. Pan, "Cloud-to-end rendering and storage management for virtual reality in experimental education," Virtual Reality & Intelligent Hardware, Vol.2, No.4, pp.368-380, 2020. https://doi.org/10.1016/j.vrih.2020.07.001
  23. W. Zou, S. Feng, X. Mao, F. Yang, and Z. Ma, "Enhancing quality of experience for cloud virtual reality gaming: An object-aware video encoding," 2021 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), pp.1-6, 2021.
  24. Unity Technologies, Unity [Internet], https://unity.com/