How Computers Work
cache
small fast memory near the CPU
Source board
This definition comes from the board below. Open it to see the term in context.
More terms from this board
- cache linethe ~64B unit moved between levels
- DRAMmain memory (RAM)
- hit ratefraction of hits
- hit/missfound in cache / had to go slower
- latency vs bandwidthper-access delay vs throughput
- LRU/associativitythe cache's eviction policy
- memory wallCPU-vs-RAM speed gap
- MMU/page tablethe address translator
- NUMAmemory closer to some cores than others (many-socket)
- prefetchCPU guesses & preloads the next line
- spatial localityuse nearby data
- swap/pagingspilling RAM to disk when full
- temporal localityreuse soon
- TLBa cache of translations
- virtual memoryper-process address space
- working setthe memory a program touches often