How to measure a computer performance?

How to measure a computer performance?

Computer basics

·

2 min read

The speed of a processor is dictated by the pulse frequency produced by the clock, measured in cycles per second, or Hertz (Hz).

The rate of pulses is known as the clock rate, or clock speed.

One increment, or pulse, of the clock is referred to as a clock cycle, or a clock tick.

The time between pulses is the cycle time.

A 1-GHz processor receives 1 billion pulses per second.

The number of machine instructions executed for a program until it runs to completion or for some defined time interval is known as Instruction count ( \(I_c\)).

The number of clock cycles required varies for different types of instructions such as load, store, branch etc. for any given processor. The cycle per instruction, $$ CPI = \frac{\sum_1^n (CPI_i \; \times \; I_i)}{I_c} $$ where

\(I_i\;\)- number of executed instructions of type i

The processor time \(T\) needed to execute a given program can be expressed as $$ T = I_c \; \times \; CPI \; \times \tau $$

where \(\tau\) - a constant cycle time, \(\;\tau\;=\;\frac{1}{f}\)

The four system attributes affect the performance of processor time. They are

  • Instruction set architecture
  • Compiler technology
  • Processor implementation
  • Cache & memory hierarchy

The rate at which the instructions are executed in the processor is measured as its performance, expressed as millions of instructions per second (MIPS). $$ \begin{align} MIPS\; rate & = \frac{I_c}{T\;\times \;10^6} \\\\ & = \frac{f}{CPI\; \times \; 10^6} \end{align} $$ where \(f\) - frequency of processor clock.


Did you find this article valuable?

Support The Data Ilm by becoming a sponsor. Any amount is appreciated!