기말 TF 대비 리뷰
Performance
- Factors that affect performance
- Algorithm
- Programming language, compiler, architecture
- Processor and memory system
- I/O system and OS
- Moore's law
- IC capacity doubles every 18-24 months
- IC capacity doubles every 18-24 months
- Response time ( = latency, execution time)
- How long it takes for a task to be completed
- How long it takes for a task to be completed
- Throughput
- Total work done per unit time (ex. tasks / hr)
- Total work done per unit time (ex. tasks / hr)
- Comparing performance
$$Performance_{X} = \frac{1}{Execution\,Time_{X}}$$
$$\frac{Performance_{X}}{Performance_{Y}} = \frac{Execution\,Time_{Y}}{Execution\,Time_{X}}$$ - Measuring execution time
- Elapsed time: total response time, including all aspects (processing, I/O, overhead, idle time)
- CPU time: time spent processing a given job. Comprises user CPU time and system CPU time
- $$CPU\,Time = (CPU\,Clock\,Cycles) * (Clock\,Cycle\,Time) = \frac{CPU\,Clock\,Cycles}{Clock\,Rate}$$
- Our focus = user CPU time
- CPU clocking
- Clock frequency (Clock rate): cycles per second
- Clock period: duration of a clock cycle (= reciprocal of clock frequency)
- Instruction count, CPI, and performance
- Instruction count (IC): determined by program, ISA and compiler
- Average cycles per instruction (CPI): determined by CPU hardware, instruction mix
- Performance revisited
$$Time = Cycles * Clock\,Period = \frac{IC * Avg.\,CPI}{Clock\,Rate}$$
'컴퓨터 구조' 카테고리의 다른 글
[컴퓨터 구조] 챕터5 리뷰 (3) | 2024.12.20 |
---|---|
[컴퓨터 구조] 챕터 4리뷰 (0) | 2024.11.15 |