140
click to vote
PPOPP
2010
ACM
16 years 5 days ago
2010
ACM
Loop vectorization, a key feature exploited to obtain high performance on Single Instruction Multiple Data (SIMD) vector architectures, is significantly hindered by irregular memo...