Orin-NX Cuda-mini-nbody docker testing on Ubuntu 20.04 via the Phoronix Test Suite. MAXN Default: Processor: ARMv8 Cortex-A78E @ 1.98GHz (8 Cores), Motherboard: EDK II 3.1-32827747, Memory: 16GB, Disk: 512GB TS512GMTE300S, Graphics: EFI VGA, Network: Realtek RTL8111/8168/8411 + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.10.104-tegra (aarch64), Display Driver: NVIDIA, OpenGL: 4.6.0, Compiler: GCC 9.4.0 + CUDA 11.4, File-System: overlayfs, Screen Resolution: 1920x1080, System Layer: docker CUDA Mini-Nbody 2015-11-10 Test: Flush Denormals To Zero (NBody^2)/s > Higher Is Better MAXN Default . 41.63 |========================================================= CUDA Mini-Nbody 2015-11-10 Test: SOA Data Layout (NBody^2)/s > Higher Is Better MAXN Default . 39.42 |========================================================= CUDA Mini-Nbody 2015-11-10 Test: Cache Blocking (NBody^2)/s > Higher Is Better MAXN Default . 56.49 |========================================================= CUDA Mini-Nbody 2015-11-10 Test: Loop Unrolling (NBody^2)/s > Higher Is Better MAXN Default . 58.41 |========================================================= CUDA Mini-Nbody 2015-11-10 Test: Original (NBody^2)/s > Higher Is Better MAXN Default . 37.17 |=========================================================