Browse > Article
http://dx.doi.org/10.3745/KIPSTA.2005.12A.6.557

Load Balancing of Unidirectional Dual-link CC-NUMA System Using Dynamic Routing Method  

Suh Hyo-Joon (카톨릭대학교 컴퓨터정보공학부)
Abstract
Throughput and latency of interconnection network are important factors of the performance of multiprocessor systems. The dual-link CC-NUMA architecture using point-to-point unidirectional link is one of the popular structures in high-end commercial systems. In terms of optimal path between nodes, several paths exist with the optimal hop count by its native multi-path structure. Furthermore, transaction latency between nodes is affected by congestion of links on the transaction path. Hence the transaction latency may get worse if the transactions make a hot spot on some links. In this paper, I propose a dynamic transaction routing algorithm that maintains the balanced link utilization with the optimal path length, and I compare the performance with the fixed path method on the dual-link CC-NUMA systems. By the proposed method, the link competition is alleviated by the real-time path selection, and consequently, dynamic transaction algorithm shows a better performance. The program-driven simulation results show $1{\~}10\%$ improved fluctuation of link utilization, $1{\~}3\%$ enhanced acquirement of link, and $1{\~}6\%$ improved system performance.
Keywords
Dual-link Interconnection Network CC-NUMA System; Transaction Path; Dynamic Routing; Load Balancing;
Citations & Related Records
Times Cited By KSCI : 2  (Citation Analysis)
연도 인용수 순위
1 J. Torrellas and V. Krishnan, 'A Chip-Multiprocessor Architecture with Speculative Multithreading,' IEEE Trans. Comp., Sept., 1999   DOI   ScienceOn
2 S.C.Woo, M.Ohara, E.Torrie, J.P.Singh, and A.Gupta. 'Methodological considerations and characterization of the SPLASH-2 parallel application suite,' Proc. International Symp. on Computer Architecture, pp.24-36, 1995
3 L. Barroso and M. Dubois, 'The Performance of CacheCoherent Ring-based Multiprocessors,' Proc. of the 20th International Symp, on Computer Architecture, pp.268-277, May, 1993
4 서효중, '다중 프로그램 환경에 적합한 이중 연결 CC-NUMA시스템', 한국정보처리학회논문지A, Vol.11, No.3, pp.321-328, 2004. 06   과학기술학회마을
5 A-T. Nguyen, M. Michael, A. Sharma, and J. Torrellaz, 'The Augmint multiprocessor simulation toolkit for Intel x86 architecture,' Proc. of the IEEE Conf. Computer Design, Oct., 1996
6 Hyo-Joong Suh, Sung Woo Chung, 'DRACO: Optimized CC-NUMA system with Novel Dual-Link Interconnections to Reduce the Memory Latency,' Proc. MEDIA-2004 Workshop, July, 2004
7 서효중, '건너뜀 이중링크를 갖는 고확장성 CC- NUMA시스템', 한국정보과학회논문지A, Vol.31 , No.9, pp.487-494, 2004. 10   과학기술학회마을
8 http://panda.snu.ac.kr/nrl/
9 P.Mak, G.E.Strait, M.A.Blake, K.W.Kark, V.K.Papazova, A.E.Seigler, G.A.Van Huben, L.Wang, and G.C.Wellwood, 'Processor subsystem interconnect architecture for a large symmetric multiprocessor system,' IBM Journal, Vol.48. No.3/4, May/July, 2004   DOI
10 http://www.dg.com/
11 Tom Lovett and Russel Clapp, 'STiNG: A CC-NUMA Computer System for the Commercial Marketplace,' Proc. of the 23th International Symp, on Computer Architecture, pp. 308-317, May, 1996   DOI
12 Daniel Lenoski, James Laudon, Kourosh Gharachorloo, Wolf-Dietrich Weber, Anoop Gupta, John Hennessy, Mark Horowitz, and Monica S. Lam, 'The Stanford Dash multiprocessor,' IEEE Trans. Comp., Vol.25 No.3, pp.63-79, Mar., 1992   DOI   ScienceOn
13 IEEE Computer Society, IEEE Standard for Scalable Coherent Interface(SCI), Institute of Electrical and Electronics Engineers, Aug., 1993
14 S. S. Nemawarkar, R. Govindarajan, G. R. Gao, and V.K. Agarwal 'Performance of Interconnection Network in Multithreaded Architectures,' Lecture Notes in Computer Science, Vol.817, pp.823-826, 1994