Time Sliced Optical Burst Switching - PowerPoint PPT Presentation

1 / 20
About This Presentation
Title:

Time Sliced Optical Burst Switching

Description:

Applied Research Laboratory, Washington University in St. Louis, Missouri, USA. ... applied research laboratory. 6. Effects of burst length on performance ... – PowerPoint PPT presentation

Number of Views:69
Avg rating:3.0/5.0
Slides: 21
Provided by: JAI6
Category:

less

Transcript and Presenter's Notes

Title: Time Sliced Optical Burst Switching


1
Time Sliced Optical Burst Switching
  • Jeyashankher Ramamirtham, Jonathan Turner
  • jai_at_arl.wustl.edu, jst_at_arl.wustl.edu,
  • Applied Research Laboratory,
  • Washington University in St. Louis,
  • Missouri, USA.

2
Motivation
  • Optical burst switching switches data in the
    wavelength domain to provide acceptable
    statistical multiplexing performance
  • Wavelength conversion is a crucial building block
  • Almost all wavelength conversion techniques use
    equivalent of laser and optical modulator
  • Contribute to 30-40 of electronic router cost
  • Time Sliced Optical Burst Switching (TSOBS) is
    designed to eliminate need for wavelength
    conversion by switching in the time domain
    instead
  • Can be done with very little buffering capacity

3
Network architecture
WDM links
Concentrator
Packet from a host
One or more wavelengths used to transmit control
information (Burst Header Cells).
Concentrators transmit user data in time-division
channels. May aggregate user packets to improve
efficiency.
Space-division optical switches switch data from
incoming timeslots to timeslots in the outgoing
link (possibly delaying the data)
Lower bit-rate host interface (e.g. Gig-Ethernet)
Frame of time slots
4
Design Issues Timeslot duration
  • Timeslot duration
  • each timeslot has a guard time to allow for
    timing uncertainties
  • solid-state switches perform switching in 10 ns
    or less
  • accuracy of synchronization of timeslots also is
    a determining factor for the guard time
  • guard times of 10-100 ns implies that we need to
    have a timeslot of the order of 1 ?s for data
    transmission efficiency
  • at transmission rates of 10 Gb/s, 1 ?s timeslot
    corresponds to approximately 1100 bytes of user
    data
  • with 90 timeslots per frame, each timeslot
    corresponds to a 100 Mb/s channel

5
Design Issues Timeslots per frame
  • For single timeslot bursts, good performance with
    moderate number of timeslots per frame

6
Effects of burst length on performance
  • Performance reduces if bursts are longer than a
    single timeslot
  • We expect most packets to be contained within a
    single timeslot

7
Design Issues Signal degradation
  • Optical signals degrade when traveling through
    multiple hops requiring regeneration midway
  • Equip each switch with few ports of regenerators
  • BHC of burst carries information on the number of
    hops, distance traveled
  • Information in BHC used to regenerate bursts as
    necessary
  • If bursts travel through ten or more routers
    before regeneration, TSOBS has a decisive cost
    advantage
  • Minimizing number of switching operations within
    a switch becomes very important

8
Switch architecture
lh
lh
The controller used the information in the BHCs
to make switching decisions and generates the
corresponding control signals
OTSI
SYNC
Optical Crossbar
Optical Crossbar
l1
l1
WDM Links
. . .
. . .
OTSI
SYNC
Crossbars perform the required space switching
Optical Time Slot Interchangers provide the
required time domain switching
BHCs
control signals
SYNC blocks synchronize incoming frame boundaries
to local timing reference using variable delay
lines, with feedback control from controller
Controller
WDM multiplexor
9
Optical Time Slot Interchanger
Signals de-multiplexed before switching and
re-multiplexed onto the delay lines. Cost of
delay lines shared by the different wavelengths.
One crossbar per wavelength to switch the signals
onto delay lines
  • Cost of crossbars is critical
  • Need to minimize the number of delay lines

10
Non-blocking OTSIs
  • Straightforward design with N delay lines of one
    timeslot each
  • Least possible delay line size N
  • Large crossbar size (N1)(N1)
  • Up to N switching operations
  • Reduce switching operations by using delay lines
    of length 1, 2,,N instead.
  • Delay lines of length 1,2,, ?N1/2?-1 and ?N1/2?,
    2?N1/2?,,(?N1/2?-1) ?N1/2?
  • Crossbar size (2?N1/2?-1)(2?N1/2?-1) 3131,
    N256
  • Length of fiber N?N1/2?/2 (2048, N 256)
  • Maximum number of switching operations 3

11
Blocking OTSIs
  • Lower complexity alternative with a small
    non-zero blocking probability
  • Natural choice of delays 1,2,,N/2
  • Crossbar size log2Nlog2N (88 for N256)
  • Length of fiber N-1 (255 for N 256)
  • Define a search procedure to find sequence of
    delay lines to switch signals onto without
    creating conflicts
  • We show that the number of switching operations ?
    3, under most conditions
  • Also, the impact of blocking on the statistical
    multiplexing performance is small

12
Search procedure
Number of timeslots/frame
u0
0
1
2
3
4
5
6
7
output
u1
u2
u4
delay 1
delay 2
delay 4
u5
u3
u6
u7
Schedule array
A square is colored if the delay line is busy
at the time
Shortest path tree
  • Schedule array implicitly defines a search graph
  • Shortest path tree is constructed
  • Node with the least number of switching
    operations chosen

13
Performance of TSOBS router
  • Simulation setup
  • Number of input and output links 16
  • Uniform random traffic with binomially
    distributed arrivals and one timeslot bursts
  • Metrics
  • Burst discard probability
  • Average number of switching operations
  • The fraction of bursts that require more than k
    switching operations 1 F(k), where k is the
    number of switching operations

14
Different no. of timeslots per frame
non-blocking
  • We do not lose much by using blocking OTSIs by
    way of performance

15
Different no. of timeslots per frame
  • For loads up to 70, average remains below 2 and
    for loads up to 90, the average remains below 3

16
Different no. of timeslots per frame
  • For N64, less than 45 of bursts require more
    than 2 switching operations
  • Less than 0.5 require more than 3 switching
    operations

17
Upper bound on no. of switching ops.
6 reduction when number of switching operations
restricted to 3 as compared to 4
  • In most cases, it suffices to restrict the number
    of switching operations down to three.

18
Varying the number of delay lines
  • Reducing number of delay lines from 6 to 4
    implies reducing the amount of fiber by a factor
    4
  • Having 4 delay lines is comparable to having 6
    delay lines in terms of performance

19
Varying the number of delay lines
  • At high loads of 90 and above, the number of
    switching operations required by the 4 delay line
    configuration is high

20
Conclusions
  • Time Sliced Optical Burst Switching
  • replace switching in wavelength domain with
    switching in time domain
  • eliminates the need for wavelength conversion
  • Presented design issues involved with TSOBS
  • moderate number time slots per frame (64-128) can
    achieve good statistical multiplexing performance
  • addressed issues with signal degradation
  • Optical Time Slot Interchanger designs
  • presented a novel non-blocking design
  • low complexity blocking designs perform as well
    as non-blocking designs
Write a Comment
User Comments (0)
About PowerShow.com