DRAGON
Dynamic Resource Allocation via GMPLS
Optical Networks
Tom LehmanUniversity of Southern CaliforniaInformation Sciences Institute (USC/ISI)
National Science Foundation
Jerry SobieskiUniversity of Maryland (UMD)Mid-Atlantic Crossroads (MAX)
Bijan JabbariGeorge Mason University (GMU)
DRAGON Team Members
• University of Maryland (UMD) Mid-Atlantic CrossRoads (MAX)
• University of Southern California Information Sciences Institute (USC/ISI)
• George Mason University (GMU) • Movaz Networks• MIT Haystack Observatory • NASA Goddard Space Flight Center (GSFC)• US Naval Observatory• National Center for Supercomputing
Applications (NCSA) Alliance Center for Collaboration, Education, Science, and Software (ACCESS)
DRAGON Objectives
• Experiment with next generation regional optical network architectures, features, capabilities
• Experiment with eScience applications– What network features and capabilities are
needed to support eScience applications?– What features do eScience applications need
to include, to best utilize next generation networks?
– Build collaborations between network community and eScience communitieso to utilize next generation networks to
enable advanced science in those domains
DRAGON Activities
• Instantiation of an Experimental Regional Optical Network in Washington D.C. region– “Hybrid” Packet Switched and Circuit Switched
Infrastructure– All optical core– Protocol agnostic (HDTV, ethernet, sonet,
fibreChannel, any optical signal)– Dynamic provisioning of end-to-end paths– Inter-Domain– Authentication, Authorization, Accounting– Scheduling
• Integrate eScience applications– eVLBI– High Definition format collaboration and remote
steering/display of visualization resources
End to End GMPLS TransportWhat is missing?
DRAGON Architecture Components
• Network Aware Resource Broker (NARB)– Inter-domain routing for GMPLS TE Capabilities– IGP/EGP Listener– Path Computation– AAA– Scheduling (and monitoring/enforcement)– Application Request Processing
• Virtual Label Switched Router (VLSR)– Proxy for non-GMPLS capable network segments
• Application Specific Topology Descriptions Language (ASTDL)– Language for application requests to network
• All Optical End-to-End Routing
VLSR Abstraction
Application Specific Topology Description Language - ASTDL
Heterogeneous Network TechnologiesComplex End to End Paths
DRAGON NetworkOptical Transport layer - Year 3
All Optical Core
Dynamic Provisioning of “Application Topologies”
DRAGON Network – Example Topology
Commercial PartnerMovaz Networks
• MEMS-based switching fabric• 400 x 400 wavelength switching, scalable to 1000s x 1000s • 9.23"x7.47"x3.28" in size • Integrated multiplexing and demultiplexing, eliminating the cost and
challenge of complex fiber management
• Dynamic power equalization (<1 dB uniformity), eliminating the need for expensive external equalizers
• Ingress and egress fiber channel monitoring outputs to provide sub-microsecond monitoring of channel performance using the OPM
• Switch times < 5ms
eVLBI Experiment Configuration - Goals
• electronic-Very Long Baseline Interferometry (e-VLBI)– MIT Haystack– NASA GSFC (GGAO)– USNO– Radio Telescopes reachable via other Infrastructures
• eVLBI Experiment Configuration
CLPK
GWU
ARLG
NASAGSFC
MAXUMD
BossNetUSNO
Radio Telescopeat GGAO
CorrelatorCorrelator
Radio Telescopesreachable via Abilene
Radio Telescopeat Haystack
Radio Telescopesreachable via NLR
MITHaystack
NLR
Abilene
ECK
Uncompressed HDTV-over-IPCurrent Method
Low latency High Definition CollaborationDRAGON Enabled
• End-to-end native SMPTE 292M transport• Media devices are directly integrated into the DRAGON environment via proxy hosts
– Register the media device (camera, display, …)– Sink and source signaling protocols– Provide Authentication, authorization and accounting.
Low Latency Visual Area Networking
• Directly share output of visualization systems across high performance networks.• DRAGON allows minimum latency paths.