nbody -ftz=true cudaMemcpyToSymbol cudaGraphicsMapResources cudaEventRecord cudaStreamQuery cudaEventCreate cudaGraphicsResourceGetMappedPointer cudaGetDeviceCount cudaEventElapsedTime cudaDeviceSynchronize cudaEventSynchronize cudaGraphicsResourceSetMapFlags cudaSetDeviceFlags cudaEventDestroy cudaDeviceCanAccessPeer cudaSetDevice cudaGraphicsUnmapResources cudaGetDeviceProperties cudaGetDevice " to the command line will allow users to set # of bodies for simulation. Adding “-numdevices=” to the command line option will cause the sample to use N devices (if available) for simulation. In this mode, the position and velocity data for all bodies are read from system memory using “zero copy” rather than from device memory. For a small number of devices (4 or fewer) and a large enough number of bodies, bandwidth is not a bottleneck so we can achieve strong scaling across these devices.]]> whole ./galaxy_20K.bin ./CMakeLists.txt ./ ../ ../../../Common Graphics Interop Data Parallel Algorithms Physically-Based Simulation CUDA GPGPU n-body simulation astrophysics opengl openGL GLU GL GLUT GLEW glut X11 ../../../common/lib/linux/x86_64 ../../../common/lib/linux/armv7l ../../../common/lib/darwin true nbody.cpp -benchmark -compare -cpu X11 GL 2:Graphics Interop 1:CUDA Advanced Topics 1:Data-Parallel Algorithms 3:Physically-Based Simulation sm35 sm37 sm50 sm52 sm53 sm60 sm61 sm70 sm72 sm75 sm80 sm86 sm87 x86_64 linux windows7 x86_64 macosx arm sbsa all exe doc/nbody_gems3_ch31.pdf