Cache Management and Designs

Question 1

Temporal locality

Answer 1

Programs exhibit temporal locality, if an item in memory is referenced, it is likely to be referenced again soon

Question 2

Spatial locality

Answer 2

Programs exhibit spatial locality, if an item in memory is referenced, then it is likely that those nearby will be referenced soon. When a specific memory block is referenced, the cache will also bring in additional memory blocks located at similar addresses

Question 3

Cache hit/Miss

Answer 3

When memory is requested from the cache it can either result in a hit or a miss

Hit - the requested data is in cache and can be read
Miss - the data is read from memory into cache first

Question 4

Direct-mapped cache + pros and cons

Answer 4

A memory address is mapped to exactly one cache address

Pros:
Easy to search

Cons:
Inefficient use of cache space

Question 5

Fully associative cache + pros and cons

Answer 5

A memory address is mapped to the least recently used cache address

Pros:
Flexible, allows the cache to be filled with any combination of memory addresses

Cons:
takes a long time to search

Question 6

Write hit - write back

Answer 6

Write into cache memory only, means faster writes but need to remember to later write to memory before replacing modified cache data

Question 7

Write hit - write through

Answer 7

Write to both cache and memory, slower but means that cache and memory are in sync

Question 8

Write miss - write around (no write allocate)

Answer 8

Write into memory only, avoids unnecessary cache access and pollution but does not utilise temporal locality

Question 9

Write miss - write allocate

Answer 9

Write into memory and read into cache - exploits temporal locality but can mean there’s unnecessary cache access and pollution