Chapter_01 of COD

8 Great Ideas in Computer Architecture

• Design for Moore’s Law
  面向摩尔定理的设计
• Use abstraction to simplify disign
  使用抽象简化设计
• Make the common case fast
  加速经常性事件
• Performance via parallelism
  通过并行提高性能
• Performance via pipelining
  通过流水线提高性能
• Performance via prediction
  通过预测提高性能
• Hierarchy of memories
  存储层次
• Dependability via redundancy
  通过冗余提高可靠性

计算机组成

• 数据通路：进行算数逻辑运算
• 控制器：在数据通路、存储器、I/O之间调度工作流
• 存储器：包括静态随机存储器和动态随机存储器
• 输入设备：如键盘
• 输出设备：如显示器

Performance

Performance = 1 / Execution Time

“X is n time faster than Y” => 性能反比于执行时间
Performance of x / Performance of Y = Execution time of Y / Execution time of X

Execution Time

Elapsed Time（实际时间）

Total response time, including all aspects

• processing
• I/O
• OS overhead
• idle time
  Determines system performance

CPU Time

影响因素：

• 算法：IC,CPI
• 编程语言：IC,CPI
• 编译器：IC,CPI
• 指令集：IC,CPI,T

计算：

• CPU时间
  • 总时间 = IC * CPI * T = IC * CPI / Rate
• 平均CPI
  • 总cycle数 = CPI * IC 加权求和
  • 平均CPI = 总cycle数 / IC = CPI加权求和
    Time spent processing a given job
• Discounts I/O time, other job’s shares
  Comprises user CPU time and system CPU time
  Different programs are affected diffrently by CPU and system performance

CPU Time = CPU Clock Cycles * Clock Cycle Time = CPU Clock Cycles / Clock Rate

Performance improved by

• Reducing number of clock cycles
• Increasing clock rate
• Hardware designer must often trade off clock rate against cycle count（许多技术再减少时钟周期的个数时，会增加时钟周期长度）

Clock Cycles = Instruction Count * Cycles Per Instruction

CPU Time = IC * CPI * Clock Cycle Time = IC * CPI / Clock Rate

Pitfall：Amdahl’s Law(阿姆达尔定律)

#阿姆达尔定律