Model Taken for Experiment: Virus
on a Network
( NetLogo Models Library : Sample Models/Networks)
Stonedahl, F.
and Wilensky, U. (2008). NetLogo Virus
on a Network model.
http://ccl.northwestern.edu/netlogo/models/VirusonaNetwork. Center for Connected Learning and
Computer-Based Modeling, Northwestern University, Evanston, IL.
§ Tool Used: NetLogo 5.2.1
(NetLego: * Wilensky, U. (1999). NetLogo. http://ccl.northwestern.edu/netlogo/. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.)
(NetLego: * Wilensky, U. (1999). NetLogo. http://ccl.northwestern.edu/netlogo/. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.)
Brief
Description About The Model:
This model demonstrates the spread of a virus
through a network. Although the model is
somewhat abstract, one interpretation is that each node represents a computer,
and we are modeling the progress of a computer virus (or worm) through this
network. Each node may be in one of
three states: susceptible, infected, or
resistant. In the academic literature
such a model is sometimes referred to as an SIR model for epidemics.
Here is what I have tried, noticed and learned:
A.
THESE ARE THE THINGS SUGGESTED TO MODIFY AND NOTICE
At the end of the run, after the virus has died out, some nodes
are still susceptible, while others have become immune.
1.
What is the ratio of the number of immune
nodes to the number of susceptible nodes?
2.
How is this affected by changing the
AVERAGE-NODE-DEGREE of the network?
THINGS AS OBSERVED AND LEARNED:
1.
1.
AND I:R
3 1:9
6 5:6
9 3:2
12 2:1
15 3:1
2.
What it is found from this experiment that It is much more difficult to quarantine a virus in a highly connected graph, because it is easy for the virus to route around paths that go dead when nodes become immune. Thus in a highly connected graph it is necessary that a greater proportion of nodes become immune before the simulation will end.
What it is found from this experiment that It is much more difficult to quarantine a virus in a highly connected graph, because it is easy for the virus to route around paths that go dead when nodes become immune. Thus in a highly connected graph it is necessary that a greater proportion of nodes become immune before the simulation will end.
B.
THESE ARE THE THINGS SUGGESTED TO MODIFY AND TRY
1.
Set GAIN-RESISTANCE-CHANCE to 0%.
2.
Under what conditions will the virus still die
out?
3.
How long does it take?
4.
What conditions are required for the virus to
live?
5.
If the RECOVERY-CHANCE is bigger than 0, even
if the VIRUS-SPREAD-CHANCE is high, do you think that if you could run the
model forever, the virus could stay alive?
THINGS AS OBSERVED AND LEARNED:
The virus will die out firstly if the graph is not sufficiently connected. It appears that an infection cannot be sustained in a graph with average node degree less than 4. Death might take anywhere from several hundred to over a thousand time steps to occur, but it will occur for sure.
The virus will die out firstly if the graph is not sufficiently connected. It appears that an infection cannot be sustained in a graph with average node degree less than 4. Death might take anywhere from several hundred to over a thousand time steps to occur, but it will occur for sure.
Once the graph is sufficiently connected, the initial outbreak
must be sufficiently large for the virus to live. For example, it appears that
an initial outbreak size of 7 or larger is sustainable for 150 nodes, an
average node degree of 4, and the default infection parameters.
As long as the pool of available hosts is not decreasing (the chance
of resistance remains 0), the virus could stay alive indefinitely even given a
non-zero chance of recovery. It also seems there is a tensity between the
chance of recovery and the chance of virus spread because as it becomes more
likely that a node will recover in a given time period the virus must also
become more likely to spread to new nodes from the infected node within that
time period.