The UK e-Science Core Programme and the Grid

Abstract

This paper describes the £120M UK ‘e-Science’ (http://www.research-councils.ac.uk/ and http://www.escience-grid.org.uk) initiative and begins by defining what is meant by the term e-Science. The majority of the £120M, some £75M, is funding large-scale e-Science pilot projects in many areas of science and engineering. The infrastructure needed to support such projects must permit routine sharing of distributed and heterogeneous computational and data resources as well as supporting effective collaboration between groups of scientists. Such an infrastructure is commonly referred to as the Grid. Apart from £10M towards a Teraflop computer, the remaining funds, some £35M, constitute the e-Science ‘Core Programme’. The goal of this Core Programme is to advance the development of robust and generic Grid middleware in collaboration with industry. The key elements of the Core Programme will be outlined including details of a UK e-Science Grid testbed. The pilot e-Science projects that have so far been announced are then briefly described. These projects span a range of disciplines from particle physics and astronomy to engineering and healthcare, and illustrate the breadth of the UK e-Science Programme. In addition to these major e-Science projects, the Core Programme is funding a series of short-term e-Science demonstrators across a number of disciplines as well as projects in network traffic engineering and some international collaborative activities. We conclude with some remarks about the need to develop a data architecture for the Grid that will allow federated access to relational databases as well as flat files.

Introduction

The term ‘e-Science’ was introduced by Dr. John Taylor, Director General of Research Councils in the UK Office of Science and Technology (OST). From his previous experience as Head of Hewlett-Packard’s European Research Laboratories, and from his experience as Director General of the OST, Taylor saw that many areas of science are becoming increasingly reliant on new ways of collaborative, multidisciplinary working. The term e-Science is intended to capture these new mode of working [1]:

‘e-Science is about global collaboration in key areas of science and the next generation of infrastructure that will enable it.’

The infrastructure to enable this science revolution is generally referred to as the Grid [2]. Two examples of such e-Science projects are from particle physics and astronomy. The world-wide particle physics community is planning an exciting new series of experiments to be carried out on the new ‘Large Hadron Collider’ (LHC) experimental facility under construction at CERN in Geneva. The goal is to find signs of the Higgs boson, key to the generation of mass for both the vector bosons and the fermions of the Standard Model of the weak and electromagnetic interactions. The experimental physicists are also hoping for indications of other new types of matter such as supersymmetric particles which may shed light on the ‘dark matter’ problem of cosmology. These LHC experiments are on a scale never before seen in physics, with each experiment involving a collaboration of over 100 institutions and over 1000 physicists from Europe, USA and Japan. When operational in 2005, the LHC will generate petabytes of experimental data per year, for each experiment. This vast amount of data needs to be pre-processed and distributed for further analysis by all members of the consortia to search for signals betraying the presence of the Higgs boson or other surprises. The physicists need to put in place an LHC Grid infrastructure that will permit the transport and data mining of such distributed data sets. There are a number of funded projects in Europe (EU DataGrid [3], EU DataTag [4], UK GridPP [5] and in the USA (NSF GriPhyN [6], DOE PPDataGrid [7], NSF iVDGL [8]) in which the particle physicists are working to build a Grid that will support these needs. Our second example of e-Science is much more directly data-centric. In the UK, the astronomers are planning to create a ‘virtual observatory’ in their e-Science AstroGrid project. There are similar initiatives in the USA with the NSF NVO [9] project and the EU AVO [10] project. The goal of these projects is to provide uniform access to a federated, distributed repository of astronomical data spanning all wavelengths from radio waves to X-rays. At present, astronomical data using different wavelengths are taken with different telescopes and stored in a wide variety of formats. Their goal is to create something like a ‘data warehouse’ for astronomical data that will enable new types of studies to be performed. Again, the astronomers are considering building a Grid infrastructure to support these Virtual Observatories. Later in this article we shall describe other types of e-Science and e-Engineering problems that have more obvious interest to industry.

The vision for a layer of ‘Grid’ middleware that provides a set of core services to enable such new types of science and engineering is due to Ian Foster, Carl Kesselman and Stephen Tuecke. In their Globus project, they have developed parts of a prototype open source Grid Toolkit [11]. Their choice of the name ‘Grid’ to describe this middleware infrastructure resonates with the idea of a future in which computing resources, compute cycles and storage, as well as expensive scientific facilities and software, can be accessed on-demand like the electric power utilities of today. These ‘e-Utility’ ideas are also reminiscent with the recent trend of the Web community towards a model of ‘Web services’ advertised by brokers and consumed by applications, which have recently been brought together in the Open Grid Services Architecture (OGSA) [12].

In the next section, we outline the general structure of the UK e-Science programme and discuss the ‘existence proof’ of the NASA Information Power Grid (IPG) [13] as the closest example of a working ‘production Grid’. We also set our activity in context by listing some of the other funded Grid projects around the world. In Section 3, we describe the UK e-Science Core Programme in some detail and in Section 4 we describe the presently funded UK e-Science pilot projects in engineering and the physical sciences. We conclude with some remarks about the evolution of Grid middleware architecture and the possible take-up of the Grid by industry.