a small, fast memory located close to the CPU that stores frequently accessed data
cache memory
the technique of fetching data into the cache before the processor actually requests it
Pythia formulates hardware prefetching as this machine learning paradigm
reinforcement learning
national animal of Scotland
unicorn! 🦄
generally faster: SRAM or DRAM
SRAM
a prefetch that arrives in the cache before the actual demand request is called this
timely prefetch (R_{AT})
in Pythia, this component stores Q-values for state-action pairs
QVStore
a computer's favourite snack
microchips!
full form of SIMD
Single Instruction, Multiple Data
this undesirable effect occurs when unnecessary prefetched data occupies cache space and evicts useful data
cache pollution
reinforcement learning algorithm that Pythia uses to update its Q-values
SARSA
which came first: the chicken or the egg?
eggs! (eggs existed long before chickens!)