Automatic Thread-Block Size Adjustment for Memory-Bound BLAS Kernels on GPUs.

Daichi Mukunoki Toshiyuki Imamura Daisuke Takahashi

Published in: MCSoC (2016)

Keyphrases

bit rate
block size
external memory
inter frame
motion vectors
subband
computational complexity
main memory
memory requirements
feature selection
multiscale
data structure
object recognition
multiresolution
image features
general purpose