2. Performance

생성일: 2021년 9월 12일 오후 6:31

Performance Metrics

• Response time(Latency, execution time) : time elapsed between start and end of a program
• Throughput(처리율) : amount of work done in a fixed time
  • faster processor will improve both
  • More processors will likely only improve throughput
  • Some policies will improve throughput and worsen response time

Execution Time

• Consider a system "X" executing a fixed workload "W"
  

Speedup and Improvement Representation

System X executes a program in 10 seconds, system Y executes the same program in 15 seconds

• System X is 1.5 times faster than system Y
• The speedup of system X over system Y is 1.5 (the ratio)
• The performance improvement of X over Y is 1.5 - 1 = 0.5 = 50%
• The execution time reduction for the program, compared to Y is (15 - 10)/15 = 33%
• The execution time increase, compared to X is (15-10)/10 = 50%

Performance Equation(방정식) - 1

Performance Equation -2

방정식 1의 CPU clock cycles에 대입하면 다음과 같다.

Factors Influencing Performance

앞의 방정식에서 사용된 변수들

• Clock cycle time: manufacturing process (how fast is each transistor), how much work gets done in each pipeline stage (more on this later)
• Number of instrs: the quality of the compiler, the instruction set architecture
• CPI: the nature of each instruction and the quality of the architecture implementation

Benchmark Suites

• SPEC rating : System Performance Evaluation Corporation
  • 시스템이 baseline machine에 비해 얼마나 빠른지 명시
  • SPEC rating 600 is 1.5 times faster than SPEC rating 400

Deriving a Single Performance Number

• 어떻게 29가지의 앱의 성능을 하나의 성능 수치로 나타낼 수 있을까?
• SPEC uses arithmetic(산수의) mean(AM) : the average of each program's execution time
• Weighted arithmetic mean(가중산술평균) : the execution times of some programs are weighted to balance priorities

Common principles

• 시스템은 평상시에도 에너지를 소비한다. (leak energy)
• Frequency(cpu clock speed)은 leakage energy에는 영향 X
• 성능 개선은 일반적으로 에너지 개선으로 귀결됨
• 90-10 rule : 10% of the program accounts for 90% of execution time (10%의 프로그램이 90%의 실행시간을 차지함)
• Principle of locality : the same data/code will be used again (temporal locality), nearby data/code will be touched next (spatial locality)