Auto-tuning GEMM Kernels for a Decoupled Access/Execute Architecture Processor.
Zeng ZhaoNaijie GuYangzhao YangPublished in: CANDAR (2013)
Keyphrases
- parallel architecture
- central processor
- multi processor
- memory access
- instruction set
- access control
- industry standard
- real time
- single chip
- hardware architectures
- computation intensive
- multithreading
- software architecture
- high speed
- management system
- random access
- computer architecture
- data flow
- kernel methods
- read write
- kernel function
- feature space
- systolic array
- support vector