Transcript National Light Rail

National LambdaRail/
Florida LambdaRail/
Tallahassee Fiber Loop
TalTech Alliance Presentation
Larry Conrad
Associate VP and CIO, FSU
Chair, Florida LambdaRail, LLC
Background
Became aware of the NLR initiative the fall of 2002
Goal was to leverage depressed telecom market to
pick-up distressed assets
Initiative came out of CA and WA
They were looking at San Diego to Seattle to
Denver to Chicago to Pittsburg to D.C.
Others got involved to “nationalize” the initiative to
include D.C. to Atlanta to Dallas to San Diego
Network footprint was being finalized, but still
“pliable”
December 4, 2003
2
Background
In just 3 working days we garnered a commitment
from 6 FL universities to invest $5M over 5 years
That commitment secured extension of the NLR
to Jacksonville...
...and a seat at the table to define the NLR
Realizing none of the 6 universities were IN JAX,
so we would have to build comparable FL network
A tribute to the strategic insight of the leaders at
those 6 FL institutions to make this kind of
commitment in these difficult fiscal times!
December 4, 2003
3
Network Drivers & Motivations
Improve networking costs…”disruptive pricing!”
Control of our networking destiny
Flexibility in responding to future needs
Responsiveness to the needs of higher ed
Innovation and support for corporate research
partners
End-to-End networking
Eliminate “place” as an issue for collaboration
December 4, 2003
4
Science Drivers & Motivations
Growing demand for “application empowered” networks
Support for high performance e-science projects
High-energy physics, astronomy, earth science,
bioinformatics, environmental
Study very complex micro to macro-scale problems over
time and space
Remote access to instruments, streaming high-definition
video, specialized visualization displays, data mining,
high-performance distributed supercomputing systems
and real-time collaboration
To optimally make use of these technologies, scientists
want high-bandwidth connectivity with known and
knowable characteristics
December 4, 2003
5
Science Drivers: Examples
New CERN facility in Switzerland comes on line in 2005—
will distribute petabyte (PB) files to multiple institutions
Jefferson Lab to distribute PB files by 2006
FSU is a managing partner for the Oakridge National
Laboratory—ORNL wants 10 Gb connectivity
SURA Coastal Oceanographic Observing Platform
(SCOOP) program to distribute PB files by 2006
NSF sponsored supercomputing centers in San Diego,
CA; Champaign, IL; and Pittsburg, PA presently have 10
Gb connectivity and are moving to 40 Gb to support GRID
computing applications
Oakridge has recently been awarded an NSF GRID
computing grant to establish a new “science ultranet” with
December 4, 2003
6
10-40 Gb connectivity
Why Fiber?
Capacity needed is not otherwise affordable
Capabilities needed are not otherwise available
Flatten out the expense growth curve...
...cheaper in the long range
Provides leverage with the telco carriers
Insurance against “monopoly behavior”
Stable and predictable anchor points
Higher Ed can’t be held hostage to what industry
thinks the market wants
December 4, 2003
7
National Lambda Rail
Sparse dark fiber National footprint
Serves very high-end Experimental and Research
Applications
4 - 10Gb Wavelengths (or networks) initially
Capable of 40 10Gb wavelengths at build out
Leverage industry shakeout: distressed vendors
are very interested in selling a National footprint
Disruptive pricing
Partnership model
December 4, 2003
8
NLR Participants
Equity owners:
–
–
–
–
–
–
–
–
CENIC (CA)
PNWGP (U of WA)
Internet2
Virginia Tech
Duke
Georgia Tech
Florida LambdaRail
CIC Group—Big 10
(IU lead)
December 4, 2003
Other:
– Carnegie Mellon/PSC
(purchase services)
– NCAR/UCAR—CO
(potential)
– Texas Group
(potential, UT Austin lead)
– New Mexico
(potential, UNM lead)
9
NLR Goals and Uses
Research! (including Network Research)
Experimentation in networking and in education
Education
Support of GRID computing
Academic Medical Center driven Clinical activities
Development & implementation of network &
computing technologies not otherwise likely to be
generally available commercially as early as
needed in R&E
December 4, 2003
10
NLR footprint and physical
layer topology
SEA
POR
BOI
BOS
NYC
CHI
OGD
CLE
DEN
SVL
PIT
WDC
KAN
LAX
RAL
NAS
STR
PHO
SAN
15808 Terminal
15808 OADM
15808 Regen
Fiber route
Leased waves
MEM
OLG
ATL
DAL
JAC
Note: California (SAN-LAX-SVL) routes shown are part of CalREN; NLR is adding waves to CalREN systems. Also the CENIC SVLDecember 4, 2003
11
Sacramento (SAC) ELH route will become part of NLR SVL-SEA in exchange for a SVL-SAC LH route NLR is building (not shown here).
NLR footprint and physical
layer topology – Phase 1
SEA
POR
BOI
CHI
OGD
CLE
DEN
SVL
PIT
WDC
KAN
RAL
LAX
SAN
15808 Terminal
15808 OADM
15808 Regen
Fiber route
Leased waves
ATL
JAC
Note: California (SAN-LAX-SVL) routes shown are part of CalREN; NLR is adding waves to CalREN systems. Also the CENIC SVLDecember 4, 2003
12
Sacramento (SAC) ELH route will become part of NLR SVL-SEA in exchange for a SVL-SAC LH route NLR is building (not shown here).
NLR Partners and Costs
NLR technology partners
– Cisco Systems (electronics/optronics)
– Level 3 (fiber)
~$80 M over 5 Years
December 4, 2003
13
NLR Status
1st link from Pittsburg to Chicago brought up midNovember
Phase 1 build-out to be completed by July 2004
NLR has recently been designated the largest,
fastest scientific network in the world
NLR is negotiating with SURA on a partnership for
Phase 2 to complete the southern route back to
Atlanta
...that could come right through Tallahassee!
December 4, 2003
14
FLR: The NLR Build Out In Florida
FLR, Florida LambdaRail
FLR equity participants are all public corporation
or 501(c)(3) entities
Current members:
FAU, FIT, FIU, FSU, Nova SE,
UF, U of Miami, UCF, UWF
Expect FLR build out to be in the $20M range
over the next 5 years
FLR is a LLC with plans to seek 501(c)(3) status
December 4, 2003
15
FLR Goals and Uses
NLR connectivity
Build a high-speed FL network for education and
research
Put FL universities on an equal footing with the
best research institutions in the nation
Initial services: 10 Gb shared IP fabric with
– Commodity ISP aggregation
– Internet2 aggregation
Potential for corporate partnerships and
economic development
December 4, 2003
16
FLR Goals and Uses
For the first time, Florida has a node on the
leading edge Research and Education networking
infrastructure
– Was not true with Internet2, NSFNet, or SURANet
– Supports nominally $1B in research funding at Florida’s
research universities
December 4, 2003
17
One FLR Build Out Option
December 4, 2003
18
FLR Status
Still in start-up mode
–
–
–
–
–
Company formed
Initial equity participant solicitation closed
Will submit 501(c)(3) application shortly
Negotiating with fiber and optronics vendors
Establishing initial contracts for support services
December 4, 2003
19
FLR Timeline
Dec. 2003: Order finalized for FLR backbone fiber
Jan. 2004: Order finalized for FLR optronics
March 2004: Build and lab test Layer 2/3 transport
equipment
April 2004: Order commodity Internet and Internet2
aggregation services
July 2004: NLR Phase 1 build-out of completed
Aug. 2004: Begin testing Layer 2/3 infrastructure
Sept. 2004: Begin connecting participants to the FLR
infrastructure
Nov. 2004: Production operation
December 4, 2003
20
Tallahassee Fiber Loop
Tie together local education entities
Provide FLR/NLR connectivity
Potential participants: DOE, CCLA, FAMU,
FSU, NWRDC, TCC
Managed by NWRDC
Status: negotiating with potential providers
December 4, 2003
21
In Summary…
Many details yet to be worked out, e.g., inter-city
fiber, optronics, and metro fiber solutions
Expect members will provide connection to other
entities
For FSU, will provide a nominal 100-fold
improvement in network speed!
…which will fundamentally redefine our ability to
collaborate and compete!
December 4, 2003
22
Questions?
www.flrnet.org
www.nationallambdarail.org