directemapped

Direct-mapped cache is a cache organization in which each block of main memory maps to exactly one cache line. It is the simplest form of cache and contrasts with set-associative and fully associative caches, which allow a memory block to reside in multiple possible cache locations.

In a direct-mapped cache, the address is divided into three parts: the tag, the index, and the block offset. The index selects a single cache line. The tag stores the high-order address bits to detect whether the data in that line corresponds to the requested block. The block offset selects the specific data within the cache block. On each memory access, the cache checks the line indicated by the index; if the stored tag matches, a cache hit occurs. If the tag differs, a miss occurs, and the corresponding block is fetched from lower-level memory and stored in that line, replacing what was there previously.

Advantages of direct mapping include low hardware cost and fast access times due to the straightforward lookup and replacement logic. It provides predictable performance and is well suited for small caches in embedded systems or simple processors.

Disadvantages center on potential conflict misses. Because every memory block has only one possible cache line, multiple blocks that map to the same line compete and can evict each other frequently, reducing hit rates for certain access patterns. This makes direct-mapped caches less flexible than set-associative caches, especially for workloads with irregular or random memory access.

In practice, direct-mapped caches are often used for level 1 data caches in conjunction with larger, more associative caches at higher levels, balancing simplicity and speed with broader cache efficiency.