Chiplet和网络加速——互连时代的两大驱动力.pdf

1、远见，超越芯所未见K i w i m o o r eAgendaKiwi SoChiplet，高性能互联平台远见，超越芯所未见远见，超越芯所未见摩尔定律增加晶体管密度增加芯片面积提升芯片内传输效率提升计算效率突破存储墙增加计算规模提升 Cluster 内传输效率提升 Cluster 间传输效率降低传输负载远见，超越芯所未见摩尔定律增加晶体管密度增加芯片面积提升芯片内传输效率提升计算效率突破存储墙增加计算规模提升 Cluster 内传输效率提升 Cluster 间传输效率降低传输负载新的瓶颈：互联新的途径：Chiplet,网络加速远见，超越芯所未见高性能计算迈进Chiplet 时代，三大主流形态

2、GraceHopperLakeFieldPonte VecchioMeteor LakeSapphire Rapids Zen3+Zen2 Zen MI200MI300RX7000AWS Graviton3M1 UltraDojoBR100昇腾910远见，超越芯所未见AMD GenoaIntel Emerald Rapids Ampere SyrinAMD Bergama Intel Sapphire RapidsAWS Graviton 4远见，超越芯所未见nIntel 7nm with 2.5D Advanced PackagenCore Counts 56nThread Counts 1

3、12nTotal LLC 105 MBnPCIe 5.0/CXL1.1 Lanes Up to 128nMemory Bandwidth 307GB/snSPECrate2017_int_base(2P)990远见，超越芯所未见nIntel Golden Cove CorenLLC 1.875MBnMesh NoC Architecturen15 Cores Per Tile,14 be usednIntel 7nm with 2.5D Advanced PackagenCore Counts 56nThread Counts 112nTotal LLC 105 MBnPCIe 5.0/CXL

4、1.1 Lanes Up to 128nMemory Bandwidth 307GB/snSPECrate2017_int_base(2P)990远见，超越芯所未见nIntel Golden Cove CorenLLC 1.875MBnMesh NoC Architecturen15 Cores Per Tile,14 be usedn400Gb/s Symmetric Crypton160Gb/s Compression+n160Gb/s De-compressionn400M Load Balancing Decisions per Second nIntel 7nm with 2.5D

5、Advanced PackagenCore Counts 56nThread Counts 112nTotal LLC 105 MBnPCIe 5.0/CXL1.1 Lanes Up to 128nMemory Bandwidth 307GB/snSPECrate2017_int_base(2P)990远见，超越芯所未见nIntel Golden Cove CorenLLC 1.875MBnMesh NoC Architecturen15 Cores Per Tile,14 be usednHS D2D with 2.5D EMIIB&Si-InterposernD2D Bandwidth:5

6、00GB/snData Rate:5GT/snBump Pitch:55umnEnergy Efficiency:0.5pj/bitnPHY Latency end-to-end TX+RX:2.4nsn400Gb/s Symmetric Crypton160Gb/s Compression+n160Gb/s De-compressionn400M Load Balancing Decisions per Second nIntel 7nm with 2.5D Advanced PackagenCore Counts 56nThread Counts 112nTotal LLC 105 MBn