The OptIPuter and Other TransLight (GLIF) Projects

1
The OptIPuter and OtherTransLight (GLIF) Projects

• Maxine Brown
• Co-Principal Investigator, StarLight and
  Euro-Link/TransLight
• Project Manager, OptIPuter
• Associate Director, Electronic Visualization
  Laboratory
• University of Illinois at Chicago
• September 9, 2004

2
The OptIPuter Project Removing Bandwidth as an
Obstacle In Data Intensive Sciences

• NSF Large Information Technology Research
  Proposal
• Cal-(IT)2 and UIC Lead CampusesLarry Smarr PI
• USC, SDSU, NW, Texas AM, Univ. Amsterdam
  Partnering Campuses
• Industrial Partners
• IBM, Sun, Telcordia, Chiaro Networks, Calient,
  Glimmerglass, Lucent
• 13.5 Million Over Five Years
• Optical IP Streams From Lab Clusters to Large
  Data Objects

NIH Biomedical Informatics
NSF EarthScope and ORION
Research Network
http//ncmir.ucsd.edu/gallery.html
siovizcenter.ucsd.edu/library/gallery/shoot1/index
.shtml
3
The OptIPuter BioScience

• Improve the evolution of health care by advancing
  global collaboration for the study of large-scale
  brain image data related to a variety of
  neurological diseases
• Enable users to better control real-time remote
  instrumentation (Osaka electron microscope) and
  to more interactively manipulate multi-gigabyte
  biomedical datasets

Biomedical Informatics Research Network (BIRN)
www.nbirn.net UCSD National Center for Microscopy
and Imaging Research (NCMIR) www-ncmir.ucsd.edu UI
C Electronic Visualization Laboratory
www.evl.uic.edu/cavern/optiputer
4
The OptIPuter GeoScience

• Enable geoscientists to readily access and
  interpret large-scale geophysical and geological
  data.
• OptIPuter hardware and software immediately
  impacted researchers abilities to view
  high-resolution Mars imagery from Explorer and
  rovers Spirit and Opportunity, and 1-meter IKONOS
  imagery of the Lake Tahoe basin.

EarthScope www.earthscope.org UCSD Scripps
Institution of Oceanography (SIO)
http//siovizcenter.ucsd.edu UIC Electronic
Visualization Laboratory www.evl.uic.edu/cavern/op
tiputer
5
The OptIPuter GeoScience

• USGS leverages OptIPuter technologies to utilize
  high-resolution (0.3-meter) ortho-imagery of 133
  most-populated metropolitan areas of the United
  States in support of Homeland Security
  initiatives
• USGS looks to the OptIPuter project to provide
  leadership in developing and deploying
  next-generation affordable, interactive,
  large-scale display and Earth science analysis
  technologies

USGS Earth Resources Observation Systems (EROS)
Data Center http//edc.usgs.gov UIC Electronic
Visualization Laboratory www.evl.uic.edu/cavern/op
tiputer
6
BarriersGigabyte Science Data Objects

• Hundred Million Pixel 2D Images
• Microscopy or Telescopes
• Remote Sensing
• GigaZone 3D Objects
• Supercomputer Simulations
• Seismic or Medical Imaging
• Interactive Analysis and Visualization of Such
  Data Objects is Impossible Over Shared Best
  Effort Internet
• Deterministic Networks Enable
• Guaranteed Bandwidth (data movement)
• Guaranteed Latency (viz/collaboration, data
  analysis)
• Guaranteed Scheduling (remote instruments)
• Interactive Analysis and Visualization of Such
  High Resolution Data Objects Requires
• Scalable Visualization Displays
• Montage and Volumetric Visualization Software
• JuxtaView and Vol-a-Tile

7
OptIPuter End User Building BlocksCompute
Storage Viz Linux Clusters

• Cluster 16-128 Nodes (Typically Two Intel
  Processors)
• Storage 0.1-1 TB per Node
• Graphics Nvidia Card Per Node
• Visualization Displays Desktop, Wall, Theatre,
  Tiled, VR
• Specialized Data Source/Sink Instruments
• All Nodes Have 1 or 10 GigE I/O

Commodity GigE Switch
Fibers or Lambdas
8
OptIPuter Software Architecture from Grid to
LambdaGrid
OptIPuter Applications
Visualization
DVC 1
DVC 2
DVC 3
Layer 5 SABUL, RBUDP, Fast, GTP
Real-Time Objects
Security Models
Data Services DWTP
Higher Level Grid Services
Grid and Web Middleware (Globus/OGSA/WebServices
/J2EE)
Layer 4 XCP
Node Operating Systems
l-configuration, Net Management
Physical Resources
DVCDistributed Virtual Computer
Source Andrew Chien, UCSD OptIPuter Software
Systems Architect
9
What is the OptIPuter?

• Applications Drivers ? Interactive Large Data
  Objects
• OptIPuter Nodes ? Scalable PC Cluster LambdaGrid
  Browser
• IP over Lambda Connectivity ?Predictable
  Backplane
• Open Source LambdaGrid Middleware? Network is
  Reservable
• Data Retrieval and Mining ? Global Virtual Disk
  Drives
• High Defn. Visualization, Collaboration ?
  Ultra-Reality TV

www.optiputer.net
See Nov 2003 Communications of the ACM for
Articles on OptIPuter Technologies
10
Communications of the ACM (CACM)
Volume 46, Number 11November 2003Special
issue Blueprint for the Future of
High-Performance Networking

• Introduction, Maxine Brown (guest editor)
• TransLight a global-scale LambdaGrid for
  e-science, Tom DeFanti, Cees de Laat, Joe
  Mambretti, Kees Neggers, Bill St. Arnaud
• Transport protocols for high performance, Aaron
  Falk, Ted Faber, Joseph Bannister, Andrew Chien,
  Robert Grossman, Jason Leigh
• Data integration in a bandwidth-rich world,
  Ian Foster, Robert Grossman
• The OptIPuter, Larry Smarr, Andrew Chien, Tom
  DeFanti, Jason Leigh, Philip Papadopoulos
• Data-intensive e-science frontier research,
  Harvey Newman, Mark Ellisman, John Orcutt

http//www.acm.org/cacm
11
On-Line Microscopes in San Diego CreateVery
Large Biological Images
IBM 9M Pixels

• Laser Confocal Microscope
• High-speed online capability
• Using high-resolution 9- Megapixel IBM displays
  to prototype software
• 150 Megapixel composite image sizes need to be
  projected

Source David Lee, NCMIR, UCSD
12
Large Microscope Images Allow BothFine Detail
and Global Context
13
Large Microscope Images Allow BothFine Detail
and Global Context
Large Scale Brain Maps
14
Large Microscope Images Allow BothFine Detail
and Global Context
Large Scale Brain Maps
15
(No Transcript)
16
LambdaRAM a Networked Memory Abstraction

• Concept / Goals
• Giant pool of clustered memory to provide low
  latency access to large remote data sets by
  aggressive use of available network bandwidth.
  (Currently read-only)
• Relieve application developers from building
  their own data-prefetching middleware from
  scratch.

Visualization of the Pre-Fetch Algorithm

• Accomplishments
• Data is prefetched based on access patterns or
  prefetching algorithms.
• Integration with JuxtaView.
• Developed a visualization of the prefetching
  algorithm.
• Experiments in LAN, MAN (StarLight to UIC), WAN
  (Amsterdam to UIC) show 2-5 fold performance
  improvement over traditional memory-mapped files.

8-14
1-7
all
none
Displayed region
17
HDTV Feedback of Remote InstrumentsOptIPuter
NCMIR/BIRN researchers work with KDDI to stream
live HDTV from the worlds largest microscope in
Osaka, Japan. High-quality video is essential for
resolving useful information, such as changes in
gradients in a high-noise, low-contrast
environment. HDTV combined with dedicated lambdas
will provide lower latencies and control of
network jitter, especially important in these
large streams of video data. This step is the
first in a data acquisition feedback loop for
instrumentation steering, control, computation
and visualization.
http//ncmir.ucsd.edu
18
Distribution and Visualization of Brain Maps
OptIPuter
GridFTP is used to distribute very large (gt1Gb)
HDTV brain maps from the UCSD multi-photon
microscope to other OptIPuter on-line sites in
Southern California (UCSD, UCI, SDSU, and USC).
Data is sent from the OptIPuters IBM storage
cluster to local resources. The goal is to be
able to transfer data from an instrument to
OptIPuter-connected resources in real time,
enabling researchers to steer the data
acquisition process and monitor progress, which
can currently take as long as 22 days.
NCMIR Telescience Portal
http//ncmir.ucsd.edu, https//telescience.ucsd.ed
u
19
Todays Aerial Imaging is gt500,000 TimesMore
Detailed than Landsat7 Satellite Images
Shane DeGross
SDSU Campus
30 meter pixels
Source Eric Frost, SDSU
4 centimeter pixels
Laurie Cooper, SDSU
20
Distribution and Visualization of Earth
ScienceOptIPuter
Here, SIO scientist Debi Kilb interacts with
IKONOS satellite imagery using JuxtaView (on left
panel) and a scene file that combines the same
imagery with topography, bathymetry and seismic
images (displayed on right panel). Both datasets
are fetched over the campus OptIPuter fiber from
the UCSD storage cluster.
http//siovizcenter.ucsd.edu
21
The National LambdaRail
22
TransLight 2004 Lambdas
European lambdas to US (red) 10Gb
AmsterdamChicago 10Gb LondonChicago 10Gb
AmsterdamNYC Canadian lambdas to US
(white) 30Gb Chicago-Canada-NYC 30Gb
Chicago-Canada-Seattle US sublambdas to Europe
(grey) 6Gb ChicagoAmsterdam Japan JGN II
lambda to US (cyan) 10Gb ChicagoTokyo European
lambdas (yellow) 10Gb AmsterdamCERN 2.5Gb
PragueAmsterdam 2.5Gb StockholmAmsterdam 10Gb
LondonAmsterdam IEEAF lambdas (blue) 10Gb
NYCAmsterdam 10Gb SeattleTokyo
CAVEWave/PacificWave (purple) 10Gb
ChicagoSeattle 10Gb SeattleLASan Diego
Northern Light
UKLight
Japan
CERN
23
GLIF Global Lambda Integrated Facility

• GLIF is a collaborative initiative among
  worldwide NRENs and institutions to connect
  Lambdas through Lights
• 3rd Annual Global Lambda Grid Workshop,
  Reykjavik, Iceland, August 27, 2003 (2 Japanese)
• 4th Meeting in Nottingham England, September 2-3,
  2004

www.glif.is
24
Global Lambda Integrated FacilityWorld Map
December 2004
Predicted international Research Education
Network bandwidth, to be made available for
scheduled application and middleware research
experiments by December 2004.
www.glif.is
25
Global Lambda Integrated FacilityPredicted
Bandwidth for Scheduled Experiments, December 2004
www.glif.is
26
Global Lambda Integrated FacilityPredicted
Bandwidth for Scheduled Experiments, December 2004
www.glif.is
27
Advanced Technologies fore-Science Visualization
and Collaboration

• Jason Leigh
• Electronic Visualization Laboratory
• University of Illinois at Chicago

28
Thank You!

• TransLight planning, research, collaborations,
  and outreach efforts are made possible, in major
  part, by funding from
• National Science Foundation (NSF) awards
  SCI-9980480, SCI-9730202, CNS-9802090,
  CNS-9871058, SCI-0225642, and CNS-0115809
• State of Illinois I-WIRE Program, and major UIC
  cost sharing
• Northwestern University for providing space,
  power, fiber, engineering and management
• Pacific Wave, StarLight, National LambdaRail,
  CENIC, PNWGP, CANARIE, SURFnet, UKERNA, and IEEAF
  for Lightpaths
• DoE/Argonne National Laboratory for StarLight and
  I-WIRE network engineering and design