Please wait a minute...
J4  2013, Vol. 47 Issue (6): 957-968    DOI: 10.3785/j.issn.1008-973X.2013.06.005
    
Bypassed channels based differentiated service implementation method for network-on-chip
QUAN Li, CHENG Ai-lian, PAN Yun, DING Yong, YAN Xiao-lang
Institute of VLSI Design, Zhejiang University, Hangzhou 310027, China
Download:   PDF(0KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

An efficient quality-of-service (QoS) method for network-on-chip was proposed to meet the low latency and low jitter transmission requirements of real-time applications in MPSoC (multiprocessing system-on-chip). The method utilizes pre-applying and dynamic scheduling mechanism to appropriately make use of bypass channels in routers. It allocates resources and handles conflicts sensibly according to communication tasks’ requirements and priorities. Concretely, the mechanism for low latency transmission service uses flexible bypass and specific virtual channels to forward latency critical messages as soon as possible, while the mechanism for low jitter transmission service employes virtual channel reservation and adaptive routing algorithm to maintain successive transmission and low latency variation of jitter critical messages. Experimental results showed that the average delay of latency-critical messages was reduced by 41% and the average latency difference of burst data was reduced by 39%, compared to the conventional QoS method which only based on priority and specific virtual channel. Thus the proposed method can provide high-quality differentiated service for QoC applications.



Published: 22 November 2013
CLC:  TP 302  
Cite this article:

QUAN Li, CHENG Ai-lian, PAN Yun, DING Yong, YAN Xiao-lang. Bypassed channels based differentiated service implementation method for network-on-chip. J4, 2013, 47(6): 957-968.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2013.06.005     OR     http://www.zjujournals.com/eng/Y2013/V47/I6/957


基于旁路通道的片上网络差别型服务实现方法

针对面向实时应用的片上多处理系统对片上网络提出的低延迟和低抖动通信质量要求,提出一种高效的QoS实现方法.该方法对于路由器中的旁路通道采用预先申请和动态调度机制,根据任务的通信需求和优先级制定了合理的资源分配和冲突处理规则.其中低延迟服务机制利用旁路和专用虚通道使延迟敏感消息尽快转发,低抖动服务机制利用虚通道保留技术和自适应路由算法维持突发消息的传输连续性.实验结果表明:与仅基于优先级和专用虚通道的QoS方法相比,该QoS方法使延迟敏感消息的平均延迟降低了41%,突发消息的平均延迟差异降低了39%,能为片上系统应用提供高质量的差别型服务.

[1] DALLY W J, TOWLES B. Route packets, not wires: on-chip interconnection networks [C]∥ Proceeding of the 38th Conference on Design Automation. Las Vegas, USA: ACM, 2001: 684-689.

[2] LI Bin, ZHAO Li, IYER R, et al. CoQoS: Coordinating QoS-aware shared resources in NoC-based SoCs [J]. Journal of Parallel and Distributed Computing, 2011, 71(5): 700-713.

[3] PASRICHA S, DUTT N. On-Chip Communication architectures: system on chip interconnect [M]. Burlington, USA: Morgan Kaufmann Publishers, 2008.

[4] RADULESCU A, DIELISSEN J, PESTANA S, et al. An efficient on-chip NI offering guaranteed services, shared-memory abstraction, and flexible network configuration [J]. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2005, 24(1): 4-16.

[5] MILLBERG M, NILSSON E, THID R, et al. Guaranteed bandwidth using looped containers in temporally disjoint networks within the nostrum network on chip [C]∥ Design, Automation and Test in Europe, 2004. Paris, France: IEEE Computer Society, 2004: 890-895.

[6] BJERREGAARD T, SPARSO J. A router architecture for connection-oriented service guarantees in the MANGO clockless network-on-chip [C]∥ Design, Automation and Test in Europe, 2005. Munich, Germany: IEEE Computer Society, 2005: 1226-1231.

[7] WEBER W, CHOU J, SWARBRICK I, et al. A quality-of-service mechanism for interconnection networks in system-on-chips [C]∥ Design, Automation and Test in Europe, 2005. Munich, Germany: IEEE Computer Society, 2005: 1232-1237.

[8] BOLOTIN E, CIDON I, GINOSAR R, et al. QNoC: QoS architecture and design process for network on chip [J]. Journal of Systems Architecture: the EUROMICRO Journal, 2004, 50(2): 105-128.

[9] KUMAR A, PEH L S, KUNDU P, et al. Express virtual channels: towards the ideal interconnection fabric. [C]∥ International Symposium on Computer Architecture 2007. San Diego, California, USA: ACM, 2007: 150-161.

[10] KUMAR A, KUNDU P, SINGH A P, et al. A 4.6T bits/s 3.6 GHz Single-cycle NoC Router with a Novel Switch Allocator in 65 nm CMOS [C]∥ 25th International Conference on Computer Design. Lake Tahoe, USA: IEEE, 2007: 63-70.

[11] LING Xin, CHOY C S. A Low-latency NoC Router with Lookahead Bypass [C]∥ Proceedings of 2010 IEEE International Symposium on Circuits and Systems. Paris, France: IEEE, 2010: 3981-3984.

[12] LI Zheng, WU Jie, SHANG Li, et al. Latency criticality aware On-Chip Communication [C]∥ Design, Automation and Test in Europe, 2009. Nice, France: IEEE Computer Society, 2009: 1052-1057.

[13] MURALI S, BENINI L, MICHELI G D. Mapping and physical planning of networks-on-chip architectures with quality-of-service guarantees [C]∥ Proceedings of the 2005 Asia and South Pacific Design Automation Conference. New York, USA: ACM, 2005: 27-32.

[14] PEH L S, DALLY W J. A delay model and speculative architecture for pipelined router [C]∥ Proceedings of the 7th International Symposium on High-Performance Computer Architecture. Monterrey, Mexico: IEEE Computer Society, 2001: 255-266.

[15] CHIU G M. The odd-even turn model for adaptive routing [J]. IEEE Transactions on Parallel and Distributed Systems, 2000, 11(7): 729-738.

[16] HU Jing-cao. Worm_sim Simulator-a cycle accurate simulator for Networks-on-Chip [EB/OL]. [2012-01-01]. http:∥www.ece.cmu.edu/~sld/software/worm_sim.php.

[17] MULLINS R. Netmaker [EB/OL]. [2012-01-01]. http:∥www-dyn.cl.cam.ac.uk/~rdm34/wiki/index.php?title=Main_Page.

[1] YE Xia, XIN Yuan, LIU Yong, LIU Peng. Stream Prefetcher based on MediaDSP[J]. J4, 2014, 48(2): 268-278.
[2] CAO Xiao-yang, PAN Yun, YAN Xiao-lang, HUAN Ruo-hong. Systolic structure for DCT with low area-time complexity[J]. J4, 2011, 45(4): 656-659.