CDA6530: Performance Models of Computers and Networks

Chapter 4: Using Matlab for Performance

Analysis and Simulation
 Objective
 Learn a useful tool for mathematical
analysis and simulation
 Interpreted language, easy to learn
 Use it to facilitate our simulation projects
 A good tool to plot simulation/experiment
results figures for academic papers
 More powerful than excel
 Could directly create .eps for Latex

2
 Introduction
 MatLab : Matrix Laboratory
 Numerical Computations with matrices
 Every number can be represented as matrix
 Why Matlab?
 User Friendly (GUI)
 Easy to work with

 Powerful tools for complex mathematics

 Matlab has extensive demo and tutorials

to learn by yourself
 Use help command

3
 Matlab Software Access
 all UCF in-campus computers have
student-version Matlab installed
 If you have no access to Matlab, you can
use Octave, an open-source free
software
 http://www.gnu.org/software/octave/
 The programming should be almost
identical

4
 Matrices in Matlab
 To enter a matrix
2 5 3
6 4 1
>> A = [2 5 3; 6 4 1]
>> B = [1:1.5:6; 2 3 4 5]
>> for i=1:4
for j=1:3
C(i,j)=i*j;
end
end
>> D =[]; D=[D;5]; D=[D;6;7]
>> E = zeros(4, 5)
5
 Basic Mathematical Operations

Remember that every variable can be a matrix!

Addition:
>> C = A + B

Subtraction:
>> D = A – B

Multiplication:
>> E = A * B (Matrix multiplication)
>> E = A .* B (Element wise multiplication, A and B same size)

Division:
Left Division and Right Division
>> F = A . / B (Element wise division)
>> F = A / B = A*inv(B) (A * inverse of B)
>> F = A . \ B (Element wise division)
>> F = A \ B=inv(A)*B (inverse of A * B)
6
 Generating basic matrices

Matrix with ZEROS:

>> A = zeros(m, n)

Matrix with ONES:

>> B = ones(m, n)

IDENTITY Matrix:
>> I = eye(m, n)

m  Rows
n  Columns
zeros, ones, eye  Matlab functions

7
 Obtain Information
 Size(A): return [m n]
 Length(A): length of a vector
 Length(A) = max(size(A))
 B = A(2:4,3:5)
 B is the subset of A from row 2 to row 4,
column 3 to column 5
 A(:, 2)=[]
 Delete second column

8
 Basic Matrix Functions
 Inv(A): inverse of A
 Rank(A): rank of matrix A
 A’: transpose of A
 Det(A): determinant
 V= eig(A): eigenvalue vector of A
 [V,D] = eig(A) produces matrices of eigenvalues (D)
and eigenvectors (V) of matrix A, so that A*V = V*D

9
 Random Number Generators
 Rand(m,n): matrix with each entry ~ U(0,1)
 You can use this for the programming project 1

 Randn(m,n): standard normal distribution

 You cannot use this in programming project 1
 You must use the polar method I introduced!

10
 Basic 2-D Figure Plot
 Plot(X, Y):
 Plots vector Y versus vector X
 Hold: next plot action on the same figure
 Title(‘title text here’)
 Xlabel(‘…’), ylabel(‘…’)
 Axis([XMIN XMAX YMIN YMAX])
 Legend(‘…’)
 Grid

 Example demo

11
 Elementary Math Function
 Abs(), sign()
 Sign(A) = A./abs(A)
 Sin(), cos(), asin(), acos()
 Exp(), log(), log10()
 Ceil(), floor()
 Sqrt()
 Real(), imag()

12
 Elementary Math Function
 Vector operation:
 Max(), min(): max/min element of a vector
 Mean(), median()
 Std(), var(): standard deviation and variance
 Sum(), prod(): sum/product of elements
 Sort(): sort in ascending order

13
 Save/Load Data
 Save fname
 Save all workspace data into fname.mat
 Save fname x y z
 Save(fname): when fname is a variable
 Load fname
 Load fname x y

 No error in data
 You can run simulation intermittently
 Save/load data between runs
14
 Input/Output for Text Files
 Input data file for further analysis in Matlab

Run simulation using C

matlab is slow in doing many loops

Use Matlab for post-data processing

Matrix calculation, utilize Matlab math functions

Simply use Matlab for figure ploting

Excel has constraint on data vector length (<300?)
 Functions:

[A,B…]= Textread(fname, format)

Read formated data

Use fprintf(), fscanf() similar to C

Note that variables here can be vectors/matrices

Show examples here of writing data to text file

15
 Advanced Graph
 Subplot(m, n, p)
 breaks the Figure window into an m-by-n
matrix of small axes, selects the p-th axes for
the current plot, and returns the axis handle.
 Semilogx(), semilogy(), loglog()

16
 3-D plot
 x=[0:10]; y=[0:10]; z=x’*y;
 mesh(x,y,z); figure; surf(x,y,z);

100
100
80
80
60
60
40
40
20
20
0
0 10
10 10
10 8
5 6
8
5 6 4
4 2
2 0 0
0 0

17
 M-file
 Script or function
 Scripts are m-files containing MATLAB statements

Functions are like any other m-file, but they accept arguments
 It is always recommended to name function file the same as the
function name
function A = changeSign(B)
% change sign for each element
[m,n] = size(B); A = zeros(m,n);
for i=1:m
for j=1:n
A(i,j)= -B(i,j);
end
end
return
18
 Online Tutorials
 Matlab itself contains many tutorials
 Other online tutorials:
 http://www.math.siu.edu/matlab/tutorials.html
 http://www.cs.cmu.edu/~ggordon/780/lecture
s/matlab_tutorial.pdf
 Google search “matlab tutorial ppt” to find a
lot more

19
Example on Using Matlab for
Markov Chain Steady State
Calculation

20
 Discrete-time Markov Chain transition
matrix:

 ¼ P = ¼ , ¼ [1 1 1… 1]T = 1
 ¼ (P – I) = 0, But we cannot use it directly
 Replace first column in (P-I) with [1 1..1]T to be
A, then we can solve the linear equation set by
¼ = [1 0 0 … 0] A-1
 Another way: P*P*P*P……

21
Tutorial on Matlab Simulink

22

Graphical programming language

Drag and draw line to program

Configure each object for parameters

Powerful modeling tool

Differential Equations

Physiological systems

Control systems

Transfer functions

M-file can call a simulink model

“sim fname”

Use current workspace variables

Simulation results can be saved to workspace variables

Thus can be process after simulink

23
Example: Internet Worm Propagation
dI (t) ´
= I (t) ¢[N ¡ I (t)]
dt ­
 N: vulnerable population
 ´ : worm host average scan rate
 : scanning IP space size

24
Example 2: RC Circuit

Transfer function:

25
 Save result to workspace variables

the save format is "structure with time".
 Suppose the workspace variable is X_t.
 Then:

X_t.time saves the simulation step times (vector)

X_t.signals.values saves the simulation results
(vector).
 plot(X_t.time, X_t.signals.values);

Variable step simulation or fixed step simulation:

"to workspace" use "-1" for sample time (inherited)

Then X_t.time has variable size

"to workspace" use "1" for sample time

Then each time tick has one result value

26