CUDA_Occupancy_calculator

CUDA_Occupancy_calculator - CUDA GPU Occupancy Calculator...

Info iconThis preview shows pages 1–4. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: CUDA GPU Occupancy Calculator The 1.) Select a GPU from the list (click): G80 2.) Enter your resource usage: Threads Per Block 256 Registers Per Thread 10 Shared Memory Per Block (bytes) 4096 (Don't edit anything below this line) 3.) GPU Occupancy Data is displayed here and in the graphs: Active Threads per Multiprocessor 768 Active Warps per Multiprocessor 24 Active Thread Blocks per Multiprocessor 3 Occupancy of each Multiprocessor 100% Maximum Simultaneous Blocks per GPU 48 (Note: This assumes there are at least this many blocks) Physical Limits for GPU: G80 Multiprocessors per GPU 16 Threads / Warp 32 Warps / Multiprocessor 24 Threads / Multiprocessor 768 Thread Blocks / Multiprocessor 8 Total # of 32-bit registers / Multiprocessor 8192 Shared Memory / Multiprocessor (bytes) 16384 Allocation Per Thread Block Warps 8 Registers 2560 Shared Memory 4096 These data are used in computing the occupancy data in blue Maximum Thread Blocks Per Multiprocessor Blocks Limited by Max Warps / Multiprocessor 3 Limited by Registers / Multiprocessor 3 Limited by Shared Memory / Multiprocessor 4 Thread Block Limit Per Multiprocessor is the minimum of these 3 CUDA Occupancy Calculator Version: 1.2 Click He For m Just follow steps 1, 2, and 3 below! (or click here for help) (Help) (Help) (Help) Copyright and License 16 80 6 12 18 24 Multiprocessor Warp Occupancy 1024 2048 6 12 18 24 Multiprocessor Warp Occupancy Threads Warps/Multiprocessor Registers 256 24 10 16 4 1 32 4 2 48 8 3 64 8 4 80 12 5 96 12 6 112 16 7 128 16 8 144 20 9 160 20 10 176 24 11 192 24 12 208 21 13 224 21 14 240 24 15 256 24 16 272 18 17 288 18 18 304 20 19 320 20 20 336 22 21 352 22 22 368 24 23 384 24 24 400 13 25 416 13 26 432 14 27 448 14 28 464 15 29 480 15 30 496 16 31 512 16 32 Your chosen resource usage is indicated by the red triangle on the graphs. e other data points represent the range of possible block sizes, register counts, and shared memory ere for detailed instructions on how to use this occupancy calculator. more information on NVIDIA CUDA, visit http://developer.nvidia.com/cuda 144 208 272 336 400 464 1 Varying Block Size Threads Per Block 4 8 12 6 12 18 24 1 Varying R Regi Multiprocessor Warp Occupancy 3072 4096 5120 6144 7168 8192 9216 10240 11264 12288 13312 14336 15360 16384 1 Varying Shared Memory Usage Shared Memory Per Thread Warps/Multiprocessor...
View Full Document

This note was uploaded on 10/06/2010 for the course CS 8803 taught by Professor Staff during the Spring '08 term at Georgia Institute of Technology.

Page1 / 19

CUDA_Occupancy_calculator - CUDA GPU Occupancy Calculator...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online