Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

yildiz presentation networks

No description
by

Fatih Yildiz

on 22 April 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of yildiz presentation networks

Questions?
Presentation
PART I

Backstory
Method 1
Method 2

PART II

Evil Secret
Multi Commodity Flow
Shortest Path
Problem
Designing a Communications Network on
Operation Area
Unleash The Power
MC
Bootstrapping
If you have this much Data,

Normal Distribution kicks in

POWER OF MORE DATA
ACO
METT-TC
M
E
T
T

T
C
SHORTEST PATH
MODIFIED SHORTEST PATH
BE MORE REALISTIC !
ANALYTICALLY MORE DATA
MEANS
MORE NODES AND MORE ARCS
MEANS
MORE TROUBLE

If you have this much Data,

Bootstrap works even better

CLT

If you have this much Data,

You better use Markov Chains

Ant Colony Optimization

Heuristic technique

Swarm Optimization

Simulation Modeling

Server Distance Matrix
LTJG FATIH YILDIZ

OA 4202

PROJECT PRESENTATION

Dynamic Enviroment
Time Layering
Commodities
Server Positions
D. Target
D. Source
msg#1
Position#1
Assumptions
A stable network infrastructure

Each message should visit Server

Node A --> Server --> Node B

Each message has 1 process-time

Server has limited capability
Problem Background
Objective : Minimum time

Network flow data
From Node A to Node B
each package 1 message
1133 Nodes (computers)
10903 Arcs (messages)

1 Server
> nCm(1133,1)
[1] 1133

> nCm(1133,2)
[1] 641278

> nCm(1133,10)
[1] 9.230783e+23
Dynamic Network

Dynamic Information Flow

Multiple Servers
Works similar to object - oriented

2 types of ants (entities)
Explorer
Worker

Only arcs store information
(You don't even have to know how many Targets)

Each step is a 1-step Markov Chain by ant
Arrive
Decide (based on arc info)
Leave

A LOT of ants
ACO
(-1)
(+1)
for msg #1
Server Positions #1
Position Node Weight Time
1 1 46 11
1 2 75 4
1 3 52 8
.
.
1 1133 33 14
2
Full transcript