What differentiates a direct-mapped cache from a fully-associative cache?

2 Direct-mapped cache

A direct-mapped cache is a cache mapping technique where each block of data from the main memory can be stored in exactly one cache line. The cache line is determined by a simple mathematical function, such as modulo division, that maps the block address to a cache line index. For example, if the cache has 16 lines, then the block address is divided by 16 and the remainder is the cache line index. This means that the cache can quickly locate the data block by applying the same function to the requested address and checking if the cache line contains a valid copy of the data. However, this also means that the cache can suffer from high conflict misses, which occur when two or more blocks that map to the same cache line are frequently accessed, causing frequent replacements and wasted cache space.

4 Trade-offs and alternatives

Direct-mapped and fully-associative caches represent two extreme cases of cache mapping techniques, and each has its own advantages and disadvantages. A direct-mapped cache is simple and fast, but prone to conflict misses. A fully-associative cache is flexible and efficient, but complex and slow. To balance these trade-offs, there are also intermediate cache mapping techniques, such as set-associative and skewed-associative caches, that combine the features of both direct-mapped and fully-associative caches. A set-associative cache divides the cache into several sets, each of which can store a fixed number of blocks, and uses a function to map the block address to a set index, and then searches the set for a match. A skewed-associative cache uses two or more functions to map the block address to different cache line indices, and then searches the cache lines corresponding to these indices for a match.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?