Textbook Terminology for Shared Memory Architectures
UMA: Uniform Memory Access architecture
- Single bus multiprocessors
- When accessing non-cache memory, all memory accesses take the same amount of time
- Non-cache memory is viewed as centrally located
- Processor caches are coherent with central memory and with each other.
NUMA: Non Uniform Memory Access architecture
- Main memory is physically distributed among nodes
- Processor caches only coherent with local memory - NOT coherent with other processors
- Access time to non-local memory depends on distance to memory location.