V22.0436 - Prof. Grishman

Lecture 23:   Cache Performance;  Input-Output Devices

Cache Performance (text, p. 477)

Calculate cache performance in terms of its effect on the CPI:
assume each miss (for instruction, data load, or data store) leads to a miss penalty, measured in clock cycles
(resulting from the CPU stalling while it waits for data from main memory)

Instruction fetch miss cycles / instruction = instruction miss rate x miss penalty
Data load/store miss cycles / instruction = % of load/store instructions x data miss rate x miss penalty
Total miss cycles / instruction = instruction fetch miss cycles/instruction + data load/store miss cycles/instruction

Effective CPI is increased by total miss cycles / instruction

Input-Output Devices

very wide range of device speeds and types (storage devices, computer-computer communication devices,  human interfaces)

Rotating magnetic storage

Solid-state disks

Input-output: requirements on the processor  / IO device interface