정보과학회 논문지 (Journal of KIISE)

  • 제42권11호
  • /
  • Pages.1332-1338
  • /
  • 2015
  • /
  • 2383-630X(pISSN)
  • /
  • 2383-6296(eISSN)
한국정보과학회 (Korean Institute of Information Scientists and Engineers)
권호 표지
DOI QR코드

DOI QR Code

묵시적 동기화 기반의 고성능 다중 GPU 렌더링

High-Performance Multi-GPU Rendering Based on Implicit Synchronization

  • 김영욱 (성균관대학교 전자전기컴퓨터공학과) ;
  • 이성길 (성균관대학교 소프트웨어학과)
  • 투고 : 2015.09.01
  • 심사 : 2015.09.15
  • 발행 : 2015.11.15
⟨ 이전 논문 다음 논문 ⟩

초록

최근 고품질, 초고해상도 실시간 렌더링 지원을 위하여 다중 GPU 렌더링에 대한 관심이 커지고 있다. 실시간 렌더링에서 여러 개의 GPU로 고성능을 달성하기 위해서는 GPU 간의 데이터 전송 지연과 프레임 합성 부하를 고려해야 한다. 이 논문은 이러한 부하를 최소화하고 다중 GPU의 효율을 향상하기 위해 split frame 렌더링의 동기화를 묵시적 질의 기반으로 향상하는 기법을 제안한다. 또한, 이러한 묵시적 동기화 기반 프레임 합성을 지원하기 위한 메시지 큐 기반의 렌더링 스케줄링 알고리즘도 제안한다. 본 알고리즘을 적용한 실험은 본 알고리즘이 기존 알고리즘 대비 200% 이상 효율을 향상함을 확인하였다.

Recently, growing attention has been paid to multi-GPU rendering to support real-time high-quality rendering at high resolution. In order to attain high performance in real-time multi-GPU rendering, great care needs to be taken to reduce the overhead of data transfer among GPUs and frame composition. This paper presents a novel multi-GPU algorithm that greatly enhances split frame rendering with implicit query-based synchronization. In order to support implicit synchronization in frame composition, we further present a message queue-based scheduling algorithm. We carried out an experiment to evaluate our algorithm, and found that our algorithm improved rendering performance up to 200% more than previously existing algorithms.

키워드

과제정보

연구 과제 주관 기관 : 한국연구재단

참고문헌

  1. J. R. Monfort, and M. Grossman, "Scaling of 3d game engine workloads on modern multi-gpu systems," Proc. of the Conference on High Performance Graphics 2009, pp. 37-46, 2009.
  2. W. Braithwaite, and S. Venkataraman, "Mixing Graphics & Compute with multi-GPU," GPU Technology Conference, 2013.
  3. L. Chen, O. Villa, and S. Krishnamoorthy, "Dynamic load balancing on single-and multi-GPU systems," IEEE International Symposium on Parallel & Distributed Processing, pp. 1-12, 2010.
  4. E. Rustico, G. Bilotta, A. Herault, C. Del Negro, and G. Gallo, "Advances in multi-GPU smoothed particle hydrodynamics simulations," IEEE Trans. on Parallel and Distributed Systems, Vol. 25, No. 1, pp. 43-52, 2014. https://doi.org/10.1109/TPDS.2012.340
  5. S. Eilemann, M. Makhinya, and R. Pajarola, "Equalizer: A scalable paralle rendering framework," IEEE Trans. on Visualization and Computer Graphics, Vol. 15, No. 3, pp. 436-452, 2009. https://doi.org/10.1109/TVCG.2008.104
  6. J. B. Hernando, J. Biddiscombe, B. Bohara, S. Eilemann, and F. Schürmann, "Practical parallel rendering of detailed neuron simulations," Eurographics Symposium on Parallel Graphics and Visualization, pp. 49-56, 2013.
  7. R. Hagan, and Y. Cao, "Multi-gpu load balancing for in-situ visualization," The 2011 International Conference on Parallel and Distributed Processing Techniques and Applications, 2011.
  8. S. Eilemann, A. Bilgili, M. Abdellah, J. Hernando, M. Makhinya, R. Pajarola, and F. Schurmann, "Parallel rendering on hybrid multi-gpu clusters," Eurographics Symposium on Parallel Graphics and Visualization, pp. 109-117, 2012.
  9. H. Liu, P. Wang, K. Wang, X. Cai, L. Zeng, and S. Li, "Scalable multi-gpu decoupled parallel rendering approach in shared memory architecture," 2011 International Conference on Virtual Reality and Visualization, pp. 172-178, 2011.
  10. S. Molnar, M. Cox, D. Ellsworth, and H. Fuchs, "A sorting classification of parallel rendering," IEEE Computer Graphics and Applications, Vol. 14, No. 4, pp. 23-32, 1994. https://doi.org/10.1109/38.291528