Which technique is used to reduce memory access latency by fetching and storing data before it is actually needed?

A$)$ Cache prefetching

B$)$ Branch prediction

C$)$ Register renaming

D$)$ Instruction$-$level parallelism