This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
KOM - Multimedia Communications LabProf. Dr.-Ing. Ralf Steinmetz (director)
Dept. of Electrical Engineering and Information TechnologyDept. of Computer Science (adjunct professor)
TUD – Technische Universität Darmstadt Merckstr. 25, D-64283 Darmstadt, Germany
PeerfactSim.KOM – A Simulator for Large Scale Peer-to-Peer Systems
Support for R&D of New P2P Applications
CeBIT – TTN Hessen | Halle 9, Stand C224 – 10. March 2008
Dipl.-Inform. Sebastian KauneDipl.-Inform. Konstantin PussepDipl.-Math., Dipl.Inf. Kalman GraffiDipl.-Ing. Aleksandra KovacevicDipl.-Wirtsch.-Ing. Nicolas Liebau
Started as a peer-to-peer system… The early Internet (late 1960s, ARPANET) Goal: share computing resources
The switch to master/slave or client/server The explosion of Internet on 1994 brought millions of people on network People needs just to send e-mail, surf the Web, buy the things
they just request the service and wait for the answer People are just temporary connected to Internet (via dial-up, DSL) Shortage of IP addresses (user-to-host name ratio ≈ 5)
First ARPANET imp log:The first message ever to be sent over the ARPANET from the UCLA SDS Sigma 7 Host computer to the SRI SDS 940 Host computer.
UCLA
UCSB
SRI
Utah
KOM – Multimedia Communications Lab 5
Example of Client-Server Communication
KOM – Multimedia Communications Lab 6
Client-Server Communication: Drawbacks
€costs
single-point-of-failure
bottleneck
scalable?
++
+
++
+
+--
--
--
-
KOM – Multimedia Communications Lab 7
Peer-to-Peer Communication
+--
+
+ --
++-
--+
+
Share resources Share costs Avoid bottlenecks
KOM – Multimedia Communications Lab 8
Research & Development of New (Peer-to-Peer) Applications?
SpecificationWhat to make?
DesignHow to make it?
ImplementationMake it!
TestingDoes it work?
Problem StatementWhat to research?
Review of the State of the Art
What is already done?
SolutionInventing!
EvaluationDoes the idea work?
RE
SE
AR
CH
DE
VE
LO
PM
EN
T
Hard, risky, slow!
KOM – Multimedia Communications Lab 9
PeerfactSim.KOM - Peer-to-Peer Systems Simulator
Application Layer
Transport Layer
Overlay Layer
User Layer
UDP TCP
Online-timeModel
Behavior
Jitter PackageLoss
Kademlia
Sim
ula
tion
En
gin
e
Chord
Network Layer RTT
KOM – Multimedia Communications Lab 10
Application
Transport
Overlay
User Sim
ulation Eng
ine
Network
PeerfactSim.KOM
Complex InternetStructure
GNP
X
Y
Z
(x1, y1, z1)
(x3, y3, z3)
(x2, y2, z2)
GeoLocation
Simple GeometricSpace
How PeerfactSim.KOM Works?
KOM - Multimedia Communications LabProf. Dr.-Ing. Ralf Steinmetz (director)
Dept. of Electrical Engineering and Information TechnologyDept. of Computer Science (adjunct professor)
TUD – Technische Universität Darmstadt Merckstr. 25, D-64283 Darmstadt, Germany
PeerfactSim.KOM – A Simulator for Large Scale Peer-to-Peer Systems
CeBIT – TTN Hessen | Halle 9, Stand C224 – 10. March 2008
Dipl.-Inform. Sebastian KauneDipl.-Inform. Konstantin PussepDipl.-Math., Dipl.Inf. Kalman GraffiDipl.-Ing. Aleksandra KovacevicDipl.-Wirtsch.-Ing. Nicolas Liebau