Abstract
Computer architecture is transiting from the multicore era into the heterogeneous era in which heterogeneous architectures use on-chip networks to access shared resources and how a network is configured will likely have a significant impact on overall performance and power consumption. Recently, heterogeneous network on chip (NoC) has been proposed not only to achieve performance comparable to that of the NoCs with buffered routers but also to reduce buffer cost and energy consumption. However, heterogeneous NoC design for heterogeneous GPU-CPU architectures has not been studied in depth. This paper first evaluates the performance and power consumption of a variety of static hot-potato based heterogeneous NoCs with different buffered and bufferless router placements, which is helpful to explore the design space for heterogeneous GPU-CPU interconnection. Then it proposes Unidirectional Flow Control (UFC), a simple credit-based flow control mechanism for heterogeneous NoC in GPU-CPU architectures to control network congestion. UFC can guarantee that there are always unoccupied entries in buffered routers to receive flits coming from adjacent bufferless routers. Our evaluations show that when compared to hot-potato routing, UFC improves performance by an average of 14.1% with energy increased by an average of 5.3% only.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Ma K, Li X, Chen W et al. GreenGPU: A holistic approach to energy efficiency in GPU-CPU heterogeneous architectures. In Proc. the 41st Int. Conf. Parallel Processing, September 2012, pp.48-57.
Lee J, Samadi M, Park Y et al. Transparent CPU-GPU collaboration for data-parallel kernels on heterogeneous systems. In Proc. the 22nd Int. Conf. Parallel Architectures and Compilation Techniques, Sept. 2013, pp.245-255.
Lee J, Kim H. TAP: A TLP-aware cache management policy for a CPU-GPU heterogeneous architecture. In Proc. the 18th Int. Symp. High Performance Computer Architecture, February 2012, pp.91-102.
Borkar S. Thousand core chips: A technology perspective. In Proc. the 44th Conf. Design Automation, June 2007, pp.746-749.
Hoskote Y, Vangal S, Singh A et al. A 5-GHz mesh interconnect for a teraflops processor. IEEE Micro, 2007, 27(5): 51-61.
Owens J D, Dally W J, Ho R et al. Research challenges for on-chip interconnection networks. IEEE Micro, 2007, 27(5): 96-108.
Wentzlaff D, Griffin P, Hoffmann H et al. On-chip interconnection architecture of the tile processor. IEEE Micro, 2007, 27(5): 15-31.
Taylor M B, Lee W, Miller J et al. Evaluation of the raw microprocessor: An exposed-wire-delay architecture for ILP and streams. ACM SIGARCH Computer Architecture News, 2004, 32(2): 2-13.
Moscibroda T, Mutlu O. A case for bufferless routing in on-chip networks. ACM SIGARCH Computer Architecture News, 2009, 37(3): 196-207.
Michelogiannakis G, Sanchez D, Dallv W J et al. Evaluating bufferless flow control for on-chip networks. In Proc. the 4th Int. Symp. Networks-on-Chip, May 2010, pp.9-16.
Jafri S A R, Hong Y J, Thottethodi M et al. Adaptive flow control for robust performance and energy. In Proc. the 43rd Int. Symp. Microarchitecture, December 2010, pp.433-444.
Nychis G P, Fallin C, Moscibroda T et al. On-chip networks from a networking perspective: Congestion and scalability in many-core interconnects. ACM SIGCOMM Computer Communication Review, 2012, 42(4): 407-418.
Fallin C, Craik C, Mutlu O. CHIPPER: A low-complexity bufferless deflection router. In Proc. the 17th Int. Symp. High Performance Computer Architecture, February 2011, pp.144-155.
Zhao H, Kandemir M, Ding W et al. Exploring heterogeneous NoC design space. In Proc. Int. Conf. Computer-Aided Design, November 2011, pp.787-793.
Nilsson E. Design and implementation of a hot-potato switch in a network on chip [Master Thesis]. Royal Institute of Technology, Sweden, 2002.
Lee J, Li S, Kim H et al. Adaptive virtual channel partitioning for network-on-chip in heterogeneous architectures. ACM Trans. Design Automation of Electronic Systems, 2013, 18(4): 48:1-48:28.
Kim H, Kim Y, Kim J. Clumsy flow control for highthroughput bufferless on-chip networks. IEEE Computer Architecture Letters, 2013, 12(2): 47-50.
Kahng A B, Li B, Peh L S et al. ORION 2.0: A power-area simulator for interconnection networks. IEEE Trans. Very Large Scale Integration Systems, 2012, 20(1): 191-196.
Henning J L. SPEC CPU2006 benchmark descriptions. ACM SIGARCH Computer Architecture News, 2006, 34(4): 1-17.
Che S, Boyer M, Meng J et al. Rodinia: A benchmark suite for heterogeneous computing. In Proc. Int. Symp. Workload Characterization, October 2009, pp.44-54.
Patil H, Cohn R, Charnev M et al. Pinpointing representative portions of large Intel® Itanium® programs with dynamic instrumentation. In Proc. the 37th Int. Symp. Microarchitecture, December 2004, pp.81-92.
Grot B, Hestness J, Keckler S W, Multu O. Express cube topologies for on-chip interconnects. In Proc. the 15th Int. Symp. High Performance Computer Architecture, February 2009, pp.163-174.
Balfour J, Dally W J, Black-Schaffer D et al. An energyefficient processor architecture for embedded systems. IEEE Computer Architecture Letters, 2008, 7(1):29-32.
Author information
Authors and Affiliations
Corresponding author
Additional information
This work was supported by the National Natural Science Foundation of China under Grant Nos. 61202076, 61202062.
Rights and permissions
About this article
Cite this article
Fang, J., Leng, ZY., Liu, ST. et al. Exploring Heterogeneous NoC Design Space in Heterogeneous GPU-CPU Architectures. J. Comput. Sci. Technol. 30, 74–83 (2015). https://doi.org/10.1007/s11390-015-1505-6
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11390-015-1505-6