Tailoring circuit-switched network-on-chip to application-specific system-on-chip by two optimization schemes

被引:10
作者
Chang, Kuel-Chung [1 ]
Shen, Jih-Sheng [1 ]
Chen, Tien-Fu [1 ]
机构
[1] Natl Chung Cheng Univ, Dept Comp Sci & Informat Engn, Chiayi 621, Taiwan
关键词
design application specific; interconnection; low power; networks on chip; systems on chips;
D O I
10.1145/1297666.1297678
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
As the number of cores on a chip increases, power consumed by the communication structures takes a significant portion of the overall power budget. In this article, we first propose a circuit-switched interconnection architecture which uses crossroad switches to construct dedicated channels dynamically between,any pairs of cores for nonhuge application-specific SoCs. The structure of the crossroad switch is simple, which can be regarded as a NoC-lite router, and we can easily construct a low-power on-chip network with these switches by a system-level design methodology. We also present the design methodology to tailor the proposed interconnection architecture to low-power structures by two proposed optimization schemes with profiled communication characteristics. The first scheme is power-aware topology construction, which can build low-power application-specific interconnection topologies. To further reduce the power consumption, we propose the second optimization scheme to predetermine the operating mode of dual-mode switches in the NoC at runtime. We evaluate several interconnection techniques, and the results show that the proposed architecture is more low-power and high-performance than others under some constraints and scale boundaries. We take multimedia applications as case studies, and experimental results show the power savings of power-aware topology approximate to 49% of the interconnection architecture. The power consumption can be further reduced approximately 25% by applying partially dedicated path mechanism.
引用
收藏
页数:31
相关论文
共 47 条
[1]  
[Anonymous], 2000, Parallel and Distributed Processing
[2]   AN OPTIMAL SHORTEST-PATH ROUTING POLICY FOR NETWORK COMPUTERS WITH REGULAR MESH-CONNECTED TOPOLOGIES [J].
BADR, HG ;
PODAR, S .
IEEE TRANSACTIONS ON COMPUTERS, 1989, 38 (10) :1362-1371
[3]   Joint application mapping interconnect synthesis techniques for embedded chip-scale multiprocessors [J].
Bambha, NK ;
Bhattacharyya, SS .
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2005, 16 (02) :99-112
[4]  
BENINI L, 2002, IEEE COMPUT, V1, P70
[5]   NoC synthesis flow for customized domain specific multiprocessor systems-on-chip [J].
Bertozzi, D ;
Jalabert, A ;
Murali, S ;
Tamhankar, R ;
Stergiou, S ;
Benini, L ;
De Micheli, G .
IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2005, 16 (02) :113-129
[6]  
BURGER D., 1997, SIMPLESCALAR TOOL SE
[7]   A low-power crossroad switch architecture and its core placement for network-on-chip [J].
Chang, KC ;
Shen, JS ;
Chen, TF .
ISLPED '05: PROCEEDINGS OF THE 2005 INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN, 2005, :375-380
[8]   Model checking with induction [J].
Chang, Kuangnan ;
Kung, David .
30TH ANNUAL INTERNATIONAL COMPUTER SOFTWARE AND APPLICATIONS CONFERENCE, VOL 1, REGULAR PAPERS/PANELS, PROCEEDINGS, 2006, :143-+
[9]   Lubricant thickness effect on tribological performance of ZDOL lubricated disks with hydrogenated overcoats [J].
Chen, CY ;
Fong, W ;
Bogy, DB ;
Bhatia, CS .
TRIBOLOGY LETTERS, 1999, 7 (01) :1-10
[10]  
Chen XN, 2003, ISLPED'03: PROCEEDINGS OF THE 2003 INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN, P90