[1] |
Http://www.top500.org.
|
[2] |
Bjerregarrd T, Mahadevan S.A survey of research and practices of networkonchip[J]. ACM Computing Surveys, 2006,38(1):4151.
|
[3] |
Duato J, Yalamanchili S, Ni L. Interconnection networks:Table 3Simulation result for Hybrid topologies and HybridY topologies表3Hybrid网络拓扑与HybridY网络拓扑的模拟实验结果节点规模拓扑结构基本延时吞吐率链路数开关单元交叉开关吞吐率/
|
|
开关单元基本延时×
|
|
开关单元64HD 8ary 2direct 1indirect33.86140.7640622561600800.000477545417864HD 8ary 2direct 3indirect57.40590.90156376828802560.0003130416532964HY 8ary 2direct 2number 1indirect35.79210.8765633842112960.000415047559364HY 8ary 2direct 2number 2indirect46.55450.89990064028801920.00031247134077256HD 16ary 2direct 1indirect34.76110.6937501024186882880.00003712649615256HD 16ary 2direct 4indirect78.71390.83593840961561612800.000053531229196256HD 16ary 2direct 2indirect50.00000.7611122048125445120.00006068627200256HY 16ary 2direct 2number 1indirect38.49440.7499951536104963200.00007146404037256HY 16ary 2direct 2number 3indirect64.41670.86718835841561610240.0000555310059311024HD 32ary 2direct 1indirect35.35210.66406240967475210880.0000088826426401024HD 32ary 2direct 5indirect96.30060.798438204807884861440.0000101375931101024HY 32ary 2direct 2number 1indirect39.50100.69062561445836811520.0000118323055941024HY 32ary 2direct 2number 4indirect83.01650.820312184327884851200.0000104065456851024HY 32ary 2direct 2number 2indirect54.87450.754688102406656020480.000011343652447 An engineering approach[M]. Calif:IEEE CS Press,1997.
|
[4] |
Roberto P, Crispín G, María G, et al. A new family of hybrid topologies for largescale interconnection networks[C]∥Proc of NCA, 2012:220227.
|
[5] |
Agarwal A, Bao L, Brown J, et al.Tile processor:Embedded multicore for networking and digital multimedia, Hot Chips[C]∥Proc of 2008 IEEE International SolidState Circuits, 2008:1.
|
[6] |
Gratz P, Kim C, McDonald R, et al. Implementation and evaluation of onchip network architectures[C]∥Proc of ICCD, 2006:477484.
|
[7] |
Taylor M B, Lee W, Amarasinghe S, et al. Scalar operand networks:Onchip interconnect for ILP in partitioned architectures[C]∥Proc of HPCA, 2003:341353.
|
[8] |
Howard J,Ruhl G, Dighe S, et al. An 80tile 1.28tflops networkonchip in 65nm cmos[C]∥Proc of ISSCC, 2007:8899.
|
[9] |
Kim J, Balfour J, Dally W J. Flattened butterfly for onchip networks[C]∥Proc of ACM, 2007:172182.
|
[10] |
Kim J, Dally W J, Towles B, et al. Microarchitecture of a highradix router[C]∥Proc of ISCA, 2005:420431.
|
[11] |
Gupta A K, Dally W J, Singh A, et al. Scalable optoelectronic network(SOENET)[C]∥Proc of HOTI, 2002:7176.
|
[12] |
Kim J, Dally W J, Steve S, et al. Costefficient Dragonfly topology for largescale systems[J]. IEEE Micro, 2009,29(1):3340.
|
[13] |
Faanes G, Bataineh A, Roweth D, et a1. Cray cascade:A scalable HPC system based on a Dragonfly network[C]∥Proc of SuperComputing, 2012:19.
|
[14] |
Rajamony R,Arimilli L B,Gildea K.PERCS:The IBM POWER7IH highperformance computing system[J]. IBM Journal of Research and Development, 2011,55(3):112.
|
[15] |
Leiserson C E. Fat trees:Universal networks for hardwareefficient supercomputing[J]. IEEE Transactions on Computers,1985,34(10):892901.
|
[16] |
Dally W J, Towles B. Principles and practices of Interconnection networks[M]. San Francisco:Morgan Kaufmann,2004.
|
[17] |
Ajima Y, Inoue T, Hiramoto S, et al. Tofu:interconnect for the K computer[J].Fujitsu Scientific&Technical,2012,48(3):280285.
|
[18] |
Gómez C, Gilabert F, Gómez M E, et al. RUFT:Simplifying the fat tree topology[C]∥Proc of ICPADS, 2008:153160.
|
[19] |
Nan J, Becker D U, Michelogiannakis G, et al. A detailed and flexible cycleaccurate networkonchip simulator[C]∥Proc of ISPASS,2012:8696.
|
[20] |
EscuderoSahuquillo J, Javier Garcia P, Quiles F J. An efficient strategy for reducing headofline blocking in Fat trees[J]. Concurrency and Computation:Practice and Experience,2011,23(17):22352248.
|
[21] |
http ://blogs.intel.com/research/terascale/ODIwhydifferent.pdf.
|