1 GPU 内存层次结构概述 2 cache结构 2.1 cache结构举例 2.2 texure L1 cache 2.3 L2 TLB结构 2.4 L1 data cache structure 2.5 shared memory 该论文主要解密了GPU中的内存层次结构,在此仅总结它的结论。 1 GPU 内存层次结构概述 按照CUDA中的描述,从软件角度看,GPU内存空间有6种类型:register, constant memor...
Dissecting GPU Memory Hierarchy through Microbenchmarking 主要关心 Fermi、Kepler、Maxwell,其他的paper从名字即可看出它的架构系列。 Fermi+Kepler+Maxwell L1 Physical Storage Sharing Kepler: The Kepler device has an L1 data cache, but it is designed for local rather than global memory access. In additio...
Chu, "Dissecting GPU Memory Hierarchy Through Micro benchmarking", IEEE Transactions on Parallel and Distributed Systems, Vol. 28, No. 1, pp. 72-86, 2017.X. Mei and X. Chu, "Dissecting gpu memory hierarchy through microbenchmarking," IEEE Transactions on Parallel and Distributed Systems, ...