OptIPuter and ENDfusion- Eliminating Bandwidth as an Obstacle in Data Intensive Sciences 21 st NORDUnet Networking Conference Reykjavik, Iceland August 26, 2003 Dr. Larry Smarr Director, California Institute for Telecommunications and Information Technologies Harry E. Gruber Professor, Dept. of Computer Science and Engineering Jacobs School of Engineering, UCSD
28
Embed
OptIPuter and ENDfusion- Eliminating Bandwidth as an Obstacle in Data Intensive Sciences 21 st NORDUnet Networking Conference Reykjavik, Iceland August.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
OptIPuter and ENDfusion-Eliminating Bandwidth as an Obstacle in
Data Intensive Sciences
21st NORDUnet Networking Conference
Reykjavik, Iceland
August 26, 2003
Dr. Larry Smarr
Director, California Institute for Telecommunications and Information Technologies
Harry E. Gruber Professor,
Dept. of Computer Science and Engineering
Jacobs School of Engineering, UCSD
AbstractThe OptIPuter is a radical distributed visualization, teleimmersion, data mining, and computing architecture. The National Science Foundation recently awarded a six-campus research consortium a five-year large Information Technology Research grant to construct working prototypes of the OptIPuter on campus, regional, national, and international scales. The OptIPuter project is driven by applications leadership from two scientific communities, the US National NSF's EarthScope and the National Institutes of Health's Biomedical Imaging Research Network (BIRN), both of which are beginning to produce a flood of large 3D data objects (e.g., 3D brain images or a SAR terrain datasets) which are stored in distributed federated data repositories. The project is led by the California Institute for Telecommunications and Information Technology and by the Electronic Visualization Laboratory at the University of Illinois at Chicago. Essentially, the OptIPuter is a "virtual metacomputer" in which the individual "processors" are widely distributed Linux PC clusters; the "backplane" is provided by Internet Protocol (IP) delivered over multiple dedicated 1-10 Gbps optical wavelengths; and, the "mass storage systems" are large distributed scientific data repositories, fed by scientific instruments as OptIPuter peripheral devices, operated in near real-time. Collaboration, visualization, and teleimmersion tools are provided on tiled mono or stereo super-high definition screens directly connected to the OptIPuter to enable distributed analysis and decision making. A new proposal called "ENDfusion: End-to-End Data Fusion in a National-Scale Urban Emergency Collaboratory" adapts and extends some of the OptIPuter concepts to support collaboratories for high resolution geographic information systems and earthquake response.
Where is Telecommunications Research Performed?A Historic Shift
Source: Bob Lucky, Telcordia/SAIC
U.S. Industry
Non-U.S. Universities
U.S. Universities
Percent Of The Papers Published IEEE Transactions On Communications
70%
85%
Cal-(IT)2– Research on the Future of the Internet
www.calit2.net
220 UC San Diego & UC Irvine FacultyWorking in Multidisciplinary Teams
With Students, Industry, and the Community
The California Institute for Telecommunications and Information Technology
Application Barrier One:Shared Internet Limits Speed of File Transfers
• NASA Earth Observation System– Over 100,000 Users
– Two Million Data Products Delivered per Year
• Measured Throughput for Data Transfers– 10-40 Mbps (May 2003) Mainly Over Abilene
– Interactive Megabyte Possible
Application Barrier Two:Gigabyte Science Data Objects
• Hundred Million Pixel 2-D Images– Microscopy or Telescopes
– Remote Sensing
• GigaZone 3-D Objects– Supercomputer Simulations
– Seismic or Medical Imaging
• Interactive Analysis and Visualization of Such Data Objects is Impossible Over Shared Internet
Very Large Biological Montage Images
• 2-Photon Laser Confocal Microscope– High Speed -
TeraFLOP Computing Enables High Resolution of 3D Flow Details
1024x1024x1024-A Billion Zone Computation of Compressible
Turbulence
This Simulation Run on Los Alamos ASCI
SGI Origin Array
U. Minn.SGI Visual Supercomputer Renders Images
VorticityLCSE, Univ of Minnesota
Removing User Networking Barriers:Global Intellectual Convergence
• SERENATE is a Strategic Study into the Evolution of European Research and Education Networking Over the Next 5-10 Years
• Some Findings– On A Multi-year Timescale, Move Towards Optical Switching
– Evolution Towards Heterogeneous NREN Networks (and GÉANT), with General Internet Use (Many-to-many) via Classical Packet Switching and:
– Specialised High-Speed Traffic (Few-to-Few) via Optical Paths? OptIPuter Project
– Even End-to-End Paths?? ENDfusion Project
Source: David Williams, CERN
From SuperComputers to SuperNetworks--Changing the Grid Design Point
• The TeraGrid is Optimized for Computing– 1024 IA-64 Nodes Linux Cluster– Assume 1 GigE per Node = 1 Terabit/s I/O– Grid Optical Connection 4x10Gig Lambdas = 40 Gigabit/s– Optical Connections are Only 4% Bisection Bandwidth
• The OptIPuter is Optimized for Bandwidth– 32 IA-64 Node Linux Cluster– Assume 1 GigE per Processor = 32 gigabit/s I/O– Grid Optical Connection 4x10GigE = 40 Gigabit/s– Optical Connections are Over 100% Bisection Bandwidth
OptIPuter LambdaGrid Global Laboratory
• NSF Large Information Technology Research Grant– $13.5 Million Over Five Years
• UCSD and UIC Lead Campuses—Larry Smarr PI– Co-PIs: Tom DeFanti, Jason Leigh, Phil Papadopoulos, Mark Ellisman– Project Manager, Maxine Brown
Science Drivers for a Radical New Net-Centric Architecture—The OptIPuter
• Data Intensive Neuro & Earth Sciences– Each Data Object is 3D and Gigabytes– Data in Distributed Federated Repository– Want to Interactively Analyze and Visualize– Need End-to-End Deterministic Networks
• OptIPuter Science Requirements – Computing PC Clusters– Communications Dedicated Lambdas– Data Large Lambda Attached Storage – Visualization Viz Clusters– Global Collaboration Multi-Scale Latencies
Goal: Punch a Hole Through the Internet Between
Researcher’s Lab and Remote Data!
What is the Best Application Usageof Routed vs. Switched Lambdas?
• OptIPuter Evaluating Both– Routers
– Chiaro– Juniper– Cisco– Force10
– Optical Switches– Calient– Glimmerglass
• UCSD Focusing on Routing Initially• UIC Focusing on Switching initially• Next Year Merge into Mixed Optical Fabric
½ Mile
SIO
SDSC
CRCA
Phys. Sci -Keck
SOM
JSOE Preuss
6th College
SDSCAnnex
Node M
Earth Sciences
SDSC
Medicine
Engineering High School
To CENIC
Collocation
Source: Phil Papadopoulos, SDSC; Greg Hidley, Cal-(IT)2