19-03-2012, 02:57 PM
ONe research agenda in Optical Burst Switching
[attachment=18559]
Introduction
E-Photon/ONe : Virtual Departments + Joint Projects
The OBS task force has been created within VD1 Core Networks
Main objective:
Promote publication of joint papers on OBS
Promote joint proposals on OBS
Writeup of joint teaching material
Changes in traffic profile
P2P file downloading vs. multimedia streaming
grid networking
Wavelength-Switched Networks
low network utilization and flexibility
Problems in Optical Packet-Switched Networks
lack of optical buffering
need for fast packet switching and header processing
Introduction – Why OBS
Main design objectives
decreasing complexity of OPS with still employed statistical multiplexing in optical domain
building a buffer-less network
user data travels transparently as an optical signal and cuts through the switches at very high rates
Solution
sending a header in order to temporary reserve a wavelength path
after that, sending an optical burst (a block of IP packets) through the network
Thanks to the great variability in the duration of bursts, the OBS can be view as lying between OPS (one-way reservation) and WS networks (two-way reservation)
Contention resolution—revisited
Burst loss possible due to statistical multiplexing
Application of OBS in high-speed metro/core networks
lost data has to be retransmitted on end-to-end basis (e.g. TCP)
very low burst loss probability required
need for highly effective contention resolution
Domains of contention resolution
Wavelength domain wavelength conversion
very effective as all WDM channels shared among all bursts
but: low burst loss probabilities only for many ls
additional schemes necessary, combinations beneficial
Time domain fiber delay lines (FDLs)
Space domain deflection/alternative routing
Segmentation only conflicting part of burst dropped
Bustification algorithms
Time-based, burst-size based or mixed-time-size based
In all cases input traffic goes through demultiplex and then burst formation queues
[attachment=18559]
Introduction
E-Photon/ONe : Virtual Departments + Joint Projects
The OBS task force has been created within VD1 Core Networks
Main objective:
Promote publication of joint papers on OBS
Promote joint proposals on OBS
Writeup of joint teaching material
Changes in traffic profile
P2P file downloading vs. multimedia streaming
grid networking
Wavelength-Switched Networks
low network utilization and flexibility
Problems in Optical Packet-Switched Networks
lack of optical buffering
need for fast packet switching and header processing
Introduction – Why OBS
Main design objectives
decreasing complexity of OPS with still employed statistical multiplexing in optical domain
building a buffer-less network
user data travels transparently as an optical signal and cuts through the switches at very high rates
Solution
sending a header in order to temporary reserve a wavelength path
after that, sending an optical burst (a block of IP packets) through the network
Thanks to the great variability in the duration of bursts, the OBS can be view as lying between OPS (one-way reservation) and WS networks (two-way reservation)
Contention resolution—revisited
Burst loss possible due to statistical multiplexing
Application of OBS in high-speed metro/core networks
lost data has to be retransmitted on end-to-end basis (e.g. TCP)
very low burst loss probability required
need for highly effective contention resolution
Domains of contention resolution
Wavelength domain wavelength conversion
very effective as all WDM channels shared among all bursts
but: low burst loss probabilities only for many ls
additional schemes necessary, combinations beneficial
Time domain fiber delay lines (FDLs)
Space domain deflection/alternative routing
Segmentation only conflicting part of burst dropped
Bustification algorithms
Time-based, burst-size based or mixed-time-size based
In all cases input traffic goes through demultiplex and then burst formation queues