Is GPGPU CCL worth it? A performance comparison between some GPU and CPU algorithms for solving connected components labeling on binary images
KTH Royal Institute of Technology, School of Computer Science and Communication
KTH, School of Computer Science and Communication, 2016
Connected component labeling (CCL) is a traditionally sequential problem that is hard to parallelize. This report aims to test the performance of solving CCL using massively parallel hardware through GPGPU. To achieve this several CCL algorithms were researched and implemented using C++ and OpenCL. The results showed an improvement of up to a factor of 2, which is insignificant when also considering memory transfer. In conclusion, performing CCL on the GPU is not worth it if the data has to first be transferred to and from the GPU.
February 26, 2017 by hgpu
