e-science

a presentation in the workshop of  Thailand E-science collaboration,  Chulalongkorn university, August 2004


“[Science is] a series of peaceful interludes punctuated by intellectually violent revolutions . . .[in which] . . . one conceptual world view is replaced by another.”

--Thomas Kuhn
The Structure of Scientific Revolutions

What is e-science

The use of collaborative computing to do science. Grid infrastructure enables high performance computing and data intensive sharing amongs scientist.

What is Grid infrastructure?

Grid computing is distributed computing.  It composed of many computing nodes connected into one system through networks, including high bandwidth i/o and storage.  The scale can be small (a few tens of computers) upto very large (thousands). It is scalable, that is, the grid can be incrementally larger and larger.  

The infrastructure comprises of hardware and layer of software that enable users to log in to one system and use it without worrying about physical location of computers and accessing programs and data without knowing where they are.

Think of the present internet as information services. Grid is information services plus computing services in a secure and scalable manner.

How Grid infrastructure help scientist?

Grid infrastructure let scientist to have computing power in a global scale and in coherent and transparent manner.  Grid infrastructure supports the collaborative work.  Huge amount of data can be available to all scientists for immediate use including all analysis, visualisation and simulation tools.

How scientist can collaborate?

Thai Grid initiative is already started.  Thai Grid is creating a national scale grid computing.  Initially it is of modest size, around 1,000 nodes.  The infrastructure needs applications.  Scientists can collaborate with Thai Grid to install their applications in the grid and experiment with the system.  Local nodes can joined with Thai Grid to achieve immediate useful connection with sudden increase in peak computing power and richly connected data sharing.

The mission of Thai Grid is to provide scientific and engineering communities the ability to solve problems that depend on use of large-scale and distributed resources.

Grid infrastructure

data grid  <currentgrids.pdf, p.4>
metacomputing grid <currentgrids.pdf, p.5>

Examples of the use of Grid infrastructure for science

<Blythe, p.2, p.12>
LIGO experiment, laser interferometer gravitational-wave observatory.  
use to detect astronomical objects such as pulsars.
two installations in Louisiana and Washington state.  Other projects: Virgo (Italy), GEO (Germany), Tama (Japan).
data collected during experiments is a collection of time series.
analysis is performed in time and Fourier domains.

UK e-science Projects

AstroGrid - enabling the Virtual Observatory
Axiope - Data Management and Data Sharing for Scientists
BioSimGrid - making large biomolecular simulations accessible
Climate Precition.Net - addressing Climatic uncertainty
Comb-e-Chem - exploiting combinatorial Chemistry
DAME - diagnostics and decision support across the Grid
Discovery Net - Information from Hight-throughput devices
e-Diamond - state of the art technology for breast screening
e-Family - access to protein sequences and structures
GENIE - integrating earth models to give a complete picture
GridPP - a Grid for Particle Physics
IXI - Medical imaging on the Grid

Conclusion

Thai Grid will grow with the popular use by scientists.  It is expected that as the usage grow, the computing nodes will be increased in number.  The most crucial aspects of E-science Thailand is that scientists find Grid to be indispensable for their research work and come to support it.

What does Thai Grid do?

The project will primarily concentrate on three core areas:

    * The first area is to build a consistent, robust and secure grid network that will attract additional computing resources.
    * The second area is to continuously improve and maintain the middleware in order to deliver a reliable service to users.
    * The third area is to attract new users from science as well as industry and ensure they receive the high standard of training and support they need.

Prabhas Chongstitvatana