 Customer
Spotlight
University of Melbourne chose
Polatis DirectLight
for use in
automated optical network
test-bed
The
university of Melbourne, in conjunction with ARC Special
Research Centre for Ultra-Broadband Information Networks
Department of Electrical and Electronic Engineering, has
recently chosen Polatis DirectLight for use in its RSVP-TE
control plane automated optical network test-bed.
The
university of Melbourne, in conjunction with ARC Special
Research Centre for Ultra-Broadband Information Networks
Department of Electrical and Electronic Engineering, has
recently chosen Polatis DirectLight™ for use in its RSVP-TE
control plane automated optical network test-bed.
All-optical switched networks have long been a vision
of the future in optical communications. The number of
all-optical switch startups founded in the past 6 years bears
testament to this vision, although many were unsuccessful, due
in large part to sub-optimal switch performance and a lack of
industry standards to successfully negotiate fiber switching
within networks.
The
Victoria Laboratory at the University of Melbourne has been
looking into new innovative routing algorithms and network
management schemes and have created an Automatically-Switched
Optical Network (ASON) test bed, utilizing RSVP-TE routing
algorithms developed at the lab. The ASON network
incorporates 6 Polatis DirectLight™ optical switch
trays, which will respond directly to commands from the
automated network control plane. As traffic on the network is
increased, the RSVP-TE algorithms determine the optimal
traffic patterns to prevent congestion, minimize latency, and
prevent packet loss. The appropriate commands are issue to the
DirectLight switches in the network without human
intervention.
Michael
Aquilina, developer of the control plane software stated, “The
Polatis optical switches give us flexibility in the types of
networks that we can build and therefore maximizes the
usefulness of our purchase.” Previous development work
relied heavily on simulations and software modeling,
while still quite useful, cannot provide the ultimate test
provided by real live networks. “We can now validate our
theory with practical experiments that we were unable to do
previously.” States Aquilina.
Back
to Newsletter Page
1 | 
