Chapter 7
Performance Analysis
Docsity.com
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Guidelines and tips
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Performance Analysis - Parallel and Distributed Computing - Lecture Slides, Slides of Parallel Computing and Programming
During the course of work of the Parallel and Distributed Computing we learn the core of the programming. The main points disucss in these lecture slides are:Performance Analysis, Parallel Programs, General Speedup Formula, Amdahl’s Law, Gustafson-Barsis’ Law, Karp-Flatt Metric, Isoefficiency Metric, Execution Time Components, Speedup Expression, Speedup Plot
Typology: Slides
2012/2013
1 / 49
This page cannot be seen from the preview
Don't miss anything!
Related documents
Partial preview of the text
Download Performance Analysis - Parallel and Distributed Computing - Lecture Slides and more Slides Parallel Computing and Programming in PDF only on Docsity!
Chapter 7
Performance Analysis
Learning Objectives
- Predict performance of parallel programs
- Understand barriers to higher performance
Speedup Formula
Parallel execution time
Sequential execution time
Speedup
Execution Time Components
- Inherently sequential computations: (n)
- sigma
- Potentially parallel computations: (n)
- phi
- Communication operations: (n,p)
- kappa
(n)/p
(n,p)
Speedup Plot
“elbowing out”
Efficiency
Processors used
Speedup Efficiency
Processors Parallelexecution time
Sequentialexecution time
Efficiency
Amdahl’s Law
n n p
n n
n n p n p
n n n p
Let f = ( n )/(( n ) + ( n )); i.e., f is the
fraction of the code which is inherently sequential
f ( 1 f )/ p
Example 1
- 95% of a program’s execution time occurs
inside a loop that can be executed in parallel.
What is the maximum speedup we should
expect from a parallel version of the program
executing on 8 CPUs?
Example 2
- 20% of a program’s execution time is spent
within inherently sequential code. What is the
limit to the speedup achievable by a parallel
version of the program?
5
- 2
1
- 2 ( 1 0. 2 )/
1 lim p (^) p
Pop Quiz
- An oceanographer gives you a serial program
and asks you how much faster it might run on
8 processors. You can only find one function
amenable to a parallel solution. Benchmarking
on a single processor reveals 80% of the
execution time is spent inside this function.
What is the best speedup a parallel version is
likely to achieve on 8 processors?
Amdahl Effect
- Typically (n) and (n,p) have lower
complexity than (n)/p
- As n increases, (n)/p dominates (n) & (n,p)
- As n increases, speedup increases
- As n increases, sequential fraction f decreases.
Illustration of Amdahl Effect
n = 100
n = 1,
n = 10,
Speedup
Processors