Platz ( vorh. Liste)
Rechner ( Hersteller)
Betreiber
Land
CPU-Kerne
GPU-Kerne
Linpack R max ( PFlops)
GFlops / W
HPCG ( TFlops)
1 (–)
Summit (IBM)
Oak Ridge National Lab (ORNL)
USA
8712 × 22 Power9, 3,07 GHz
26.136 × 80 Tesla V100
122,3
13,89
2925,75
2 (1)
Sunway TaihuLight (NRCPC)
National Supercomputing Center in Wuxi
China
40.960 × 260 ShenWei 26010, 1,45 GHz
–
93,01
6,05
480,84
3 (–)
Sierra (IBM)
Lawrence Livermore National Lab (LLNL)
USA
8640 × 22 Power9, 3,1 Ghz
17.280 × 80 Tesla V100
71,61
k. A.
1795,67
4 (2)
Tianhe-2A (NUDT), Upgrade des Tianhe-2
National Supercomputing Center in Guangzhou
China
35.584 × 12 Xeon E5-2692v2, 2,2 GHz
35.584 × 128 Matrix-2000
61,44
3,33
k. A.
5 (–)
ABCI
AIST
Japan
2176 × 20 Xeon Gold 6148, 2,4 GHz
4352 × 80 Tesla V100
19,88
12,05
k. A.
6 (3)
Piz Daint (Cray)
Centro Svizzero di Calculo Scientifico (CSCS)
Schweiz
5320 × 12 Xeon E5-2690v3, 2,6 GHz (+2862 × 18 Xeon E5-2695v4)
5320 × 56 Tesla V100
19,59
10,4
486,39
7 (5)
Titan (Cray
ORNL
USA
18.688 × 16 Opteron 6274, 2,2 GHz
18.688 × 14 Tesla K20x
17,59
2,14
322,32
8 (6)
Sequoia (IBM)
LLNL
USA
98.304 × 16 Bluegene/Q, 1,6 GHz
–
17,17
2,18
330,37
9 (7)
Trinity (Cray)
DOE/NNSA/Los Alamos National Lab (LANL)/ Sandia National Lab (SNL)
USA
18.872 × 16 Xeon E5-2698v3, 2,3 GHz + 9984 × 68 Xeon Phi 7250
–
14,14
3,68
546,12
10 (8)
Cori (Cray)
NERSC
USA
9152 × 68 Xeon Phi 7250
–
14,01
3,56
355,44