Cache

computer-architecture memory

Definition

Cache

A cache is a small, fast memory that stores copies of recently or frequently used data from a larger, slower memory.

Its purpose is to reduce the average memory access time by serving many accesses faster than main memory.

In a memory system, caches sit between the processor and main memory.

Terminology

Capacity

The capacity $C$ of a cache is the number of data bytes it can store.

Block Size

The block size $b$ is the number of bytes brought into the cache at once when a block is fetched from lower memory.

Number of Blocks

The number of blocks $B$ is the total number of cache blocks.

$$B=\frac{C}{b}$$

Degree of Associativity

The degree of associativity $N$ is the number of blocks in one set. An $N$-way set-associative cache therefore stores $N$ blocks per set.

Number of Sets

The number of sets $S$ is the number of cache sets. Each memory address maps to exactly one cache set.

$$S=\frac{B}{N}$$