virtual collaboratory: how climate research can be done collaboratively using the internet

1

Virtual Collaboratory: How Climate Research can be done Collaboratively using the Internet

U.S. – China Symposium and Workshop on Climate Variability, September 21-24, 1999, Beijing, China

Nancy N. Soreide, NOAA/PMEL

L. Charles Sun, NOAA/NODC

Presented by

Len Pietrafesa,

North Carolina State University

2

Vision

• Societal benefits through climate studies and improved climate prediction– Investing in observations, satellites, models, research– Investments in Information Technology will:

• Remove restrictions of the past

• Build a pipeline to the future

• Maximize the return on investments

• Build a Virtual Collaboratory for Climate Research

3

What is a Collaboratory?

The fusion of computers and electronic communications has the potential to dramatically enhance the output and

productivity of U. S. researchers. A major step toward realizing that potential can come from combining the interests

of the scientific community at large with those of the computer science and engineering community to create integrated, tool-oriented computing and communication

systems to support scientific collaboration. Such systems can be called "collaboratories."

From "National Collaboratories - Applying Information Technology for Scientific Research," Committee on a National Collaboratory, National Research Council. National Academy Press, Washington, D. C., 1993.

4

TAO as an example…

5

TAO set a standard…

• Data dissemination in initial project planning– Researcher involvement has assured data quality

• Benefits of data dissemination– Wide use of TAO data – Traditional research, modeling, forecasting groups– Over 200 refereed publications in past 5 years– Related disciplines, educational, administrative, public

• With recent advances in technology, we can do much more...

6

What are the elements of a Collaboratory for Climate Research?

• Data Access– Centralized, uniform, consistent access to geographically

distributed data in a common data format

• Data Visualization– Over-plotting data from distributed servers

– Interactive plotting with zoom functions, etc.

– 3D, Virtual Reality

• Collaboration tools– Full collaborative sharing of applications over the Internet

7

Collaboratory data access

• Data Access– Centralized, uniform, consistent access to

geographically distributed data in a common data format

• Realtime Observing System data

• Satellite data

• Gridded data / Model outputs

• Data & information products

• “Data Portal”– A “doorway” to climate data

8

The Data Portal: a “doorway” to climate data

• Why do we need a Data Portal?– Each Project Office provides a highly customized Web sites for

their data• but different datasets have different navigation and interface

characteristics

• so the user faces a bewildering spectrum of data access interfaces and locations

• Data Portal is single, uniform, consistent “doorway” to climate data in a common format

• User goes to a single location and sees a consistent interface

• Complements the customized data access

9

How do we build a Data Portal?

• Build on a proven prototype

• Next Generation, NOAAServer 2 prototype software– connects 5 geographically distributed data servers in Silver Spring, Boulder,

Seattle

– CORBA for network connections

– unified interactive Java graphics

– data from distributed servers are co-plotted together on the same axis on the users desktop

• Atmospheric and oceanic profiles, time series, ADCP data, global gridded data

http://www.pmel.noaa.gov/~nns/noaaserver/nodc-coads-tao.html

http://www.pmel.noaa.gov/~nns/noaaserver/coads-tao-raster.html

http://merlin/dwd/talks/mts98/unified_browse/

Prototype Data Portal (NOAAServer 2)

SeattleWA

BoulderCO

Silver Spring

MD

User issues request

Prototype Data Portal (NOAAServer 2)

SeattleWA

BoulderCO

Silver Spring

MD

Response to user

12

NOAAServer 2 Web Page

13

NOAAServer 2 Sample Plots

14

NOAAServer 2 Sample Plots

15

NOAAServer 2 Sample Plots

16

Prototype NOAAServer 2 Data Portal

NODCPMEL

JOSS

EXISTING NODES

17

Proposed Data Portal for HMR*

NODCPMEL

ODU

EXISTING NODES

PROPOSED NODES

*Hazardous Materials Response

18

Proposed Climate Data Portal

NODCPMEL

UHSLC(1st year)

EXISTING NODES

PROPOSED NODES

CO-OPS(2nd year)

19

Potential Collaboratory Partners:

NODCPMEL

JOSSUHSLC ODU

KOREA(KORDI)CHINA

(NMDIS)

TAIWAN(NCOR)

AUSTRIA(AODC)

CO-OPS

UA

AOMLFSU

NCSUNCDC

20

Other Technologies for the Collaboratory:

• Networks (100 Megabits/sec today, 10 Gigabits/sec in future)– Next Generation Internet (NGI) and Internet 2

• Visualization– Interactive Java graphics

– 3D, Virtual reality

– Immersion technology

• Collaboration tools– high-speed telecommunications systems for advanced collaboration

applications

– tele-immersion systems allow individuals at different locations to share a single virtual environment

– Use networks not airplanes for collaboration

21

Visualization Tools:

3D Visualization

and

Virtual Reality (VRML)

http://pmel.noaa.gov/visualization

http://www.pmel.noaa.gov/vrml

22

Collaboration Tools:

• Connects to geographically distributed climate data sets

• Uses NOAAServer2 technologies & NCSA’s Habanero

• Collaborative features– Interactive Java graphics

– Text and line annotations

– High-lighting of individual plots

– Display of data values on plots

– Support for multiple collaborating scientists

– Whiteboards

• Featured in national HPCC BlueBook 2000http://www.epic.noaa.gov/collab

OceanShare

23

OceanShare Collaborative Tool

http://www.epic.noaa.gov/collab

24

Collaboratory elements:Data Portal / Visualization /Full collaboration tools for applications sharing

Traditional users:ModelersForecastersResearchers

New users:EducatorsStudentsGeneral Public

Data & Information Users

Distributed data Observed data Satellite data Data and information products Model outputsVisualization

Uniform network accessin Collaboratory environment

Uniform network accessin Collaboratory environment

25

WebBrowser

JavaApplication

User

Network

CORBA*

Client Support

Java Servlet

Graphics

One or more Web Servers

TAO data support

CORBA*

Data

Observing System Server

Data

Common Object Request Broker Architecture (CORBA) is an industry standard Middleware. CORBA is used in the NOAAServer software from which this effort will leverage. Based on performance indicators, Java Remote Method Invocation (RMI), an alternative middleware, could easily be substituted for CORBA.

CORBA*

Network

Data ServerData Portal

Drifter Data support

CORBA*

Data

26

WebBrowser

JavaApplication

User

Network

CORBA*

Client Support

Java Servlet

Graphics

One or more Web Servers

Drifter Data support

CORBA*

Data

TAO data support

CORBA*

Data

Observing System Servers

Satellite data support

CORBA*

Data

Satellite Data Servers

Model data support

CORBA*

Data

Model Output Servers

Data

Gridded data support

CORBA*

Data

Gridded Data ServersCommon Object Request Broker Architecture (CORBA) is an industry standard Middleware. CORBA is used in the NOAAServer software from which this effort will leverage. Based on performance indicators, Java Remote Method Invocation (RMI), an alternative middleware, could easily be substituted for CORBA.

CORBA*

Network

Data ServersData Portal

27

Collaboratory functions

• Focal point for utilization of Information Technology for the advancement of climate programs

• Provide technical software/hardware support

• Data Portal

– Unified, central access to distributed data servers• Observing system data, model outputs, satellite data, data & information

products

• Visualization– Interactive Java graphics, 3D, virtual reality, immersion technology

• Collaboration tools – use networks not airplanes

28

Collaboratory Infrastructure

• Data Portal– Computer and networking hardware and software – Increased network bandwidth/speed– Next Generation Internet (NGI) connection

• Visualization– 3D, Virtual Reality, collaborative virtual environments– SGI workstations, CAVE, ImmersaDesk...

• Network collaboration tools• Relationships:

– Observing System Project Offices– Research community, Academia...– other Collaboratory nodes– Steering Committee

29

Climate Program Resource Allocation

Modeling and Forecasting

21%

Climate Observations

24%

Research46%

Data & Information

Dissemination9%

International Steering Committee

CollaboratoryPartner

CollaboratoryPartner

Collaboratory

Partner

Collaboratory Partners & CustomersProviders of Data & Information

Users of Data & Information

Observations&

Satellite Groups

Modeling&

ForecastingGroups

ResearchGroups

New Users Educational Administrators General Public

Structure of the Collaboratory for Climate Research