Docker GPU benchmark

docker run --rm --gpus all nvcr.io/nvidia/k8s/cuda-sample:nbody nbody -gpu -benchmark -numbodies=640000

> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation
GPU Device 0: "Pascal" with compute capability 6.1

> Compute 6.1 CUDA device: [NVIDIA GeForce GTX 1060 3GB]
number of bodies = 640000
640000 bodies, total time for 10 iterations: 29656.438 ms
= 138.115 billion interactions per second
= 2762.301 single-precision GFLOP/s at 20 flops per interaction

> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation
GPU Device 0: "Turing" with compute capability 7.5

> Compute 7.5 CUDA device: [Quadro RTX 3000]
number of bodies = 640000
640000 bodies, total time for 10 iterations: 21452.439 ms
= 190.934 billion interactions per second
= 3818.680 single-precision GFLOP/s at 20 flops per interaction

> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation
GPU Device 0: "Ampere" with compute capability 8.6

> Compute 8.6 CUDA device: [NVIDIA GeForce RTX 3060]
number of bodies = 640000
640000 bodies, total time for 10 iterations: 11055.232 ms
= 370.503 billion interactions per second
= 7410.066 single-precision GFLOP/s at 20 flops per interaction