2611
Comment: re-org for more SAN assignments
|
2103
rearrange san diagram resources
|
Deletions are marked like this. | Additions are marked like this. |
Line 35: | Line 35: |
An example of the network data format (for [[attachment:network-example.pdf|this SAN]]) is: {{{ 1 2 3 2 3 5 1 4 6 1 3 2 4 3 1.5 4 5 6 3 5 4 }}} Per row, the first term is the source node, the second term is the destination node, and the third term is the upper bound for a '''uniform''' distribution. For instance, the distribution for the `a13` arc should be ''Uniform(0,2)''. The largest numbered node in the second column may always be considered the ''terminal node'' and node 1 will always be the ''single source'' node of the network. |
<<Include(Assignments/StochasticAreaNetworks)>> |
Write a program SIM (meeting the usual requirements) that reproduces the results of the SAN on page 83. When invoked as
$ SIM B s RUNS
Your program should produce output such as:
OUTPUT :a12,a23,a34,a46: 0.5726 OUTPUT :a13,a36: 0.0181 OUTPUT :a12,a25,a56: 0.0015 OUTPUT :a12,a23,a36: 0.0945 OUTPUT :a14,46: 0.1944 OUTPUT :a13,a34,a46: 0.1189
The paths should be formatted without any intervening whitespace, beware:
OUTPUT :a12, a23, a34, a46: 0.5726
is no good
The value provided for each path is the critical path point estimate described in the book given s and RUNS, the seed and number of runs in the experiment. Report all point estimates to four decimal places.
Note that the there is no specific order to the output,1 but you must provide an entry for all possible paths.
When SIM is invoked as
$ SIM F s RUNS network.txt
Your program should read the data for an arbitrary network from the fourth parameter and produce the same tablular output as for the B scenario. The interprettation of s and RUNS is identical to the B scenario as well.
Consider the following network:
this would be represented as a simple text file:
1 2 3 2 3 5 1 4 6 1 3 2 4 3 1.5 4 5 6 3 5 4
Per row, the first term is the source node, the second term is the destination node, and the third term is the upper bound for a $Uniform(a,b)$ distribution. For instance, the distribution for the a13 arc should be $Uniform(0,2)$. The largest numbered node in the second column (over all the nodes in the network) may always be considered the terminal node and node 1 will always be the single source node of the network.
Submit Your Work
Log into the course website and submit your project archive file for grading.
Rubric
This work is worth 36 points.
Requirements |
Points |
Notes |
10 |
|
|
Experiment 'B' |
10 |
|
Experiment 'F' |
15 |
|
Correct submission |
1 |
|
Because we have sort -n -k2 at our disposal (1)