l9b - Amdahl's Law Useful for evaluating the impact of a...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Amdahl's Law Useful for evaluating the impact of a change. (A general observation.) Insight: • Improving a feature cannot improve performance beyond the use of the feature Suppose we introduce a particular enhancement that improves fraction f of execution time by factor S . Then: Execution time new = ((1- f )+ f/S ) × Execution time Example Task: complete EE/CS 314 project • Walk to lab: 15 minutes • Work in lab: 600 minutes • Food: 45 minutes • Walk home: 10 minutes • Total time = 670 minutes. Skip meals! And sleep in the lab! :-) ⇒ 10.4% improvement Observation: doesn't help with the time spent working. Other Performance Metrics Commonly used: MIPS (Millions of instructions per second) MIPS = Instruction count Execution time × 10 6 Problems with MIPS as a metric: • Instructions with different capabilities? • Programs with different instruction mixes? • Might not predict which machine is faster! MIPS As A Performance Metric Consider an optimized and unoptimized version of a program: Memory ALU Branch Total Instrs. Instrs. Instrs. Instrs....
View Full Document

{[ snackBarMessage ]}

Page1 / 14

l9b - Amdahl's Law Useful for evaluating the impact of a...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online