20180301 [學習筆記] 計算機結構 (1)

• 一、Classes of Computers
  • 1. Desktop computers
  • 2. Server computers
  • 3. Embedded computers
• 二、Understanding Performance
  • 1. Algorithm
  • 2. Programming language, compiler, architecture
  • 3. Processor and memory system
  • 4. I/O system (including OS)
• 三、Below your Program
  • 1. Application software
  • 2. System software
  • 3. Hardware
• 四、Levels of Program Code
  • 1. High-level language
  • 2. Assembly language
  • 3. Hardware representation
• 五、Defining Performance
  • 1. Response time
  • 2. Throughput
• 六、Relative Performance
• 七、Measuring Execution Time
  • 1. Elapsed time
  • 2. CPU time
  • 3. Instruction Count and CPI
• 八、Performance Summary

• 課程簡報
  • Chapter-1-Computer-Abstractions-and-Technology
• 參考資料
  • 計算機結構 - 速度的議題
  • GitBook - 組合語言與計算機組織
    • Computer Abstractions and Technology
    • Performance
    • CPU Time

一、Classes of Computers

1. Desktop computers

• General purpose, variety of software.
• Subject to cost / performance tradeoff.

2. Server computers

• Network based.
• High capacity, performance, reliability.
• Range from small servers to building sized.

3. Embedded computers

• Hidden as components of systems.
• Stringent power / performance / cost constraints.

二、Understanding Performance

1. Algorithm

• Determine number of operations executed.

2. Programming language, compiler, architecture

• Determine number of machine instructions executed per operation.

3. Processor and memory system

• Determine how fast instructions are executed.

4. I/O system (including OS)

• Determines how fast I/O operations are executed.

三、Below your Program

1. Application software

• Written in high-level language.

2. System software

• Compiler: translates HLL code to machine code.

3. Hardware

• Processor, memory, I/O controllers

四、Levels of Program Code

1. High-level language

• Level of abstraction closer to problem domain.
• Provides for productivity and portability.

2. Assembly language

• Textual representation of instrctions.

3. Hardware representation

• Binary digits (bits).
• Encoded instructions and data.

五、Defining Performance

1. Response time

• How long it takes to do a task.

2. Throughput

• Total work done per unit time.
  • e.g. tasks / transactions / … per hour

六、Relative Performance

• X is n time faster than Y

七、Measuring Execution Time

1. Elapsed time

• Total response time, including all aspects
  • Processing, I/O, OS overhead, idle time.
• Determines system performance

2. CPU time

• CPU Time Example

  

• Performance improved

  1. Reducing number of clock cycles.
  2. Increasing clock rate.
  3. Hardware designer must often trade off clock rate against cycle count.

3. Instruction Count and CPI

• Instruction Count：指令個數
• CPI：每一個指令需要執行多少個週期 (cycle per instruction)

• CPI Example

  

八、Performance Summary

• Performance depends on
  1. Algorithm: affects IC, possibly IC
  2. Programming language: affects IC, CPI
  3. Compiler: affects IC, CPI
  4. Instruction set architecture: affects IC, CPI, Tc