Welcome and Introduction

Overview of e-Infrastructure

• Mike Mineter (NeSC training team)

The concepts of the different aspects of e-Infrastructure are explained, including networks connecting resources of data and computation; Grids that enable these resources to be perceived as "virtual computers" that support collaboration. The impact that e-Infrastructure is beginning to have on research is described with emphasis on the enabling of collaboration. The talk sets the context for the remainder of the day, which explores e-Infrastructure at different scales - the campus, nationally and internationally.

Campus grids: e-Infrastructure within a University

• Mike Mineter (NeSC training team)

Many UK universities and institutes have already deployed e-Infrastructures. Their motivations, methods, problems and opportunities are summarised. Current emphasis is often on the better use of computational resources - so for example, teaching laboratories with many PCs can be used overnight as a high-throughput resource for computation. The potential for e-Infrastructure to contribute more widely is explored, for example through the sharing of data and in support for multi-disciplinary research.

The eMinerals project and the UCL Condor Pool

• Professsor John Brodholt (UCL)

The eMinerals project has been active in use and development of the UCL Condor Pool: * how has Condor benefitted the project and what results have been enabled? * current and future Condor development activity by eMinerals

UK-wide e-Infrastructure

• NeSC training team

UK e-Infrastructure comprises the networks (JANET, SuperJANET, UKLight), the National Grid Service and the supporting organisations such as the Grid Operations Support Centre and the Open Middleware Infrastructure Institute (OMII). An introduction to these is given. Related developments emerging from JISC are also summarised.

Grid Service Orchestration using the Business Process Execution Language (BPEL)

• Professor Wolfgang Emmerich (UCL)

Modern scientific applications often need to be distributed across grids. Increasingly applications rely on services, such as job submission, data transfer or data portal services. We refer to such services as grid services. While the invocation of grid services could be hard coded in theory, scientific users want to orchestrate service invocations more flexibly. In enterprise applications, the orchestration of web services is achieved using emerging orchestration standards, most notably the Business Process Execution Language (BPEL). We describe our experience in orchestrating scientific workflows using BPEL. We have gained this experience during an extensive case study that orchestrates grid services for the automation of a polymorph prediction application. Using this example, we explain the extent with which the BPEL language supports the definition of scientific workflows.We then describe the reliability, performance and scalability that can be achieved by executing a complex scientific workflow with ActiveBPEL, an industrial strength but freely available BPEL engine.

International e-Infrastructure

• NeSC training team

Many research collaborations are international. These can be empowered by an international e-infrastructure. Examples are given. The major European initiatives that integrate national initiatives in networking (GEANT), high performance computing (DEISA) and grid computing (EGEE) are described. The implications of international e-Infrastructure are summarised with reference to emerging standards and interoperability between grids that support many reseach communities.

Web Services based Environments for Computational Science on Grids

• Professor Peter Coveney and Dr Radhika Saksena (UCL)

We present two lightweight web-services based middleware schemes developed to facilitate easy access to distributed resources and grids for computational science. WEDS (Web services based environment for distributed simulations) is a WSRF-compliant hosting environment ideal for deploying applications onto "grids" limited to a single administrative domain. WEDS is written in perl and uses WSRF::Lite as its middleware. WEDS is easy to install and run in user space. The Application Hosting Environment (AHE) extends the WEDS philosophy to enable application deployment across administrative domains in the true "grid" sense. The AHE uses WSRF::Lite as its middleware and GridSAM as its interface to grid resource managers like Globus. The AHE client is written in java and designed to be lightweight with the ultimate objective of deployment on thin clients including PDAs. In our approach, the computational scientists are far-removed from the complexity of grid middleware, queuing systems, resource management, etc. and have easy access to grid resources for launching their applications .

Nurturing e-Research at UCL: Past, Present and Future

• Professors Roland Rosner and G David Price (UCL)

Closing discussion