epi tutorial - european processor initiative...2019/10/03 · homogeneous heterogeneous,...
TRANSCRIPT
![Page 1: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/1.jpg)
![Page 2: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/2.jpg)
![Page 3: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/3.jpg)
![Page 4: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/4.jpg)
High Performance
Computing
Data in Data out
Compute
• Starting from high
performance compute only,
HPC evolves towards:
• New workloads
• Massive volume of data
Analyze
New drivers Requirements Solutions
New workloads More computing performance (Ops
per second), also for simple
operations (FP16, FP8, INT…).
Energy efficiency (Ops per Watt).
Heterogeneity:
Generic processing
+ accelerators
Low power design
Massive volume
of data
Increased Bytes per Flops.
High bandwidth/low latency access
to all data.
High Bandwidth
Memories and 2.5D
integration
TERA1000 - CEA
< 10x energy efficiency
improvement every 4 years
![Page 5: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/5.jpg)
CPU
Cache
Memory
Bus
NIC(Network
InterConnect)
NoC + LLC
Cache Cache
Memory NIC
CacheCacheClose
Mem.
High
Speed L
ink
Close
Mem
Close
Mem
Close
Mem
Far
Mem.NIC
Generic processing
HW accelerator
Performance = ~frequency
Performance = ~nb cores
Performance = ~architecture
X86 cores, RISC cores, Co-pro extension, Accelerator, GPU, FPGA,
Real Time processing, Homogeneous, Heterogeneous, Data centric…
![Page 6: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/6.jpg)
homogeneous
heterogeneous, accelerated
China
US
Japan
Sierra / LLNL, 2019IBM P9 + NVidia GPU125 petaflops (peak)
(2021)Aurora / ANLIntel Xeon + Xe>1.0 exaflops (peak)
(2021)Frontier / ORNLAMD CPU + GPU~1.5 exaflops (peak)
Summit / ORNL, 2019IBM P9 + NVidia GPU200 petaflops (peak)148.6 petaflops
(2020-2021)Tianhe-3 / NUDTMatrix-3000>1.0 exaflops (peak)
(2020-2021)Fugaku / RIKENA64FX (Armv8.2+SVE)>0.5 exaflops
Tianhe-2a /NUDT, 2018Intel Xeon + Matrix-200094.97 petaflops (peak)
Tianhe-2 /NUDT, 2013Intel Xeon + KNC 33.86 petaflops (peak)
K / RIKEN, 2011SPARC64 VIIIfx11.28 petaflops (peak)10.51 petaflops
Sugon Exa-prototypeHygon CPU + DCU
NRCPC Exa-prototypeSW26010 based
Sunway TaihuLight /NRCPCSW26010125.43 petaflops (peak)
(?)Hygon CPU + DCU?
(?)??
Europe approach ?
![Page 7: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/7.jpg)
![Page 8: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/8.jpg)
* FPA : Framework Partnership Agreement
* FP8 : Framework Programmes 8 for 2014-2020, succeeding FP7 (2007-2013)
![Page 9: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/9.jpg)
1018
![Page 10: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/10.jpg)
![Page 11: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/11.jpg)
![Page 12: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/12.jpg)
![Page 13: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/13.jpg)
Security infrastructure
GPP processor chip
Power Management infrastructure
Generic
processingAccelerator
Real-time
processing
eFPGA
![Page 14: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/14.jpg)
ARM MPPA
eFPGA EPAC
HBMmemories
DDRmemories
PCIe gen5links
HSLlinks
D2D linksto adjacent chiplets
Application
Experts
Architects
+
Model and
simulation
Co-design
METHODOLOGY
COMPUTING UNITS
SOFTWARE
Linux Operating System
Programming tools &
Libraries
Low-level Software, Security, Power Management
Automotive eHPC
software support
EPI Processor and Reference Hardware
![Page 15: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/15.jpg)
ARM MPPA
eFPGA EPAC
HBMmemories
DDRmemories
PCIe gen5links
CCIXlinks
D2D linksto adjacent chiplets
![Page 16: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/16.jpg)
![Page 17: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/17.jpg)
ARM MPPA
eFPGA EPAC
HBMmemories
DDRmemories
PCIe gen5links
HSLlinks
D2D linksto adjacent chiplets
![Page 18: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/18.jpg)
STX
Bridge to GPP
Bridge to GPP
VPU
VRP
EPAC
![Page 19: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/19.jpg)
![Page 20: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/20.jpg)
![Page 21: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/21.jpg)
![Page 22: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/22.jpg)
AutomotiveSafety/security
MCU
![Page 23: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/23.jpg)
![Page 24: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/24.jpg)
![Page 25: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/25.jpg)
SIPEARL SAS
78600 Maisons-Laffitte
France
RCS Versailles Siren 851 434 365
WE ACCELERATE ACCELERATORS !!!!
Contact
Philippe NOTTON
+33180835490
R&D in Paris / Grenoble / Sophia Antipolis
![Page 26: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/26.jpg)
![Page 27: EPI Tutorial - European Processor Initiative...2019/10/03 · homogeneous heterogeneous, accelerated China US Japan Sierra / LLNL, 2019 IBM P9 + NVidia GPU 125 petaflops (peak) (2021)](https://reader034.vdocuments.mx/reader034/viewer/2022042122/5e9d2f95b705e81ac713ed5e/html5/thumbnails/27.jpg)