Fugaku System Configuration

Post-K system overview Post-K rack overview


Architecture Information: Download from https://github.com/fujitsu/A64FX

Architecture Armv8.2-A SVE (512 bit SIMD)
Core 48 cores for compute and 2/4 for OS activities
Normal: 2.0 GHz DP: 3.072 TF, SP: 6.144 TF, HP: 12.288 TF
Boost: 2.2 GHz DP: 3.3792 TF, SP: 6.7584 TF, HP: 13.5168 TF
Cache L1 64 KiB, 4 way, 230+ GB/s (load), 115+ GB/s (store)
Cache L2 CMG(NUMA): 8 MiB, 16 way
Node: 3.6+ TB/s
Core: 115+ GB/s (load), 57+ GB/s (store)
Memory HBM2 32 GiB, 1024 GB/s
Interconnect TofuD (28 Gbps x 2 lane x 10port)
I/O PCIe Gen3 x 16 lane
Technology 7nm FinFET

TofuD Interconnect

8B Put latency 0.49 - 0.54 usec
1MiB Put throughput 6.35 GB/s
ref. Yuichiro Ajima, et al. , “The Tofu Interconnect D,” IEEE Cluster 2018, 2018.

Storage System

Programming Language and Library

Compiler provided by Fujitsu Fortran2008 & Fortran2018 subset
C11 & GNU and Clang extensions
C++14 & C++17 subset and GNU and Clang extensions
OpenMP 4.5 & OpenMP 5.0 subset
Parallel Programming provided by RIKEN XcalableMP
Script provided by Fujitsu Python + Numpy + Scipy
Math Library provided by Fujitsu/RIKEN BLAS, LAPACK, ScaLAPACK, SSL II
Fujitsu SSL II
EigenEXA, Batched BLAS

The user-land software packages will be based on a Linux distribution. The gcc, LLVM, and Arm compilers will be also available.

System Software

OSRed Hat Enterprise Linux 8
McKernel (provided by RIKEN)
MPIFujitsu MPI (based on Open MPI), RIKEN MPI (based on MPICH)
File IOLLIO (provided by Fujitsu)
Application-oriented file IO libraries (provided by RIKEN)