Introduction Desktop Grid computing

Grids? Clouds? Computer infrastructure on-demand!  Computers grow older while you watch them Each second you do not use your computer is a second lost forever. You cannot "save" hard  disk space for later if you do not use it. A computer at home or in the office, costs money and uses  energy.

On the other hand - do not have enough computer power or disk space - when you need it, can be just as bad. Suppose you are a small company and you just did introduce a new computer based service that does extremely well beyond all expectations and you get thousands of new customers all accessing your system that is too small to handle it. By the time you ordered and installed new computers, your customers are already gone, dissapointed with your service.

But advanced computing techniques, with exotic names as Grid computing and Cloud computing now enable a computer infrastructure on-demand. This can be applied in science and industry.

This presentation focuses on Desktop Grid computing put in context of computer infrastructure on-demand. The concepts and most important software projects will be described, including BOINC, XtremWeb and EDGeS.

Masterclass: Porting applications to the Grid using the EDGeS Application Development Methodology

A Grid can be a powerful number crunching machine bringing the power of thousands of processors to an application developer's finger tips. Although current Grid infrastructures offer significant amount of resources to run computation and data intensive applications, some scenarios still overgrow the capabilities of existing Grids. Unfortunately, the two main types of Grids infrastructures, Service and Desktop Grid systems, were not interoperable until recently. The European EDGeS project is currently developing a bi-directional bridge connecting this diverse collection of resources.

Programming an application for a Grid is not easy. Current Grid application development efforts very often use ad-hoc approaches only when porting the applications. Developers do not follow any suggested methodology and this may result is poorly documented systems that do not fulfill user expectations. In order to avoid this trap, support application developers and provide guidelines when porting an application to the EDGeS Grid platform, the EDGeS Application Development Methodology (EADM) has been specified.

The masterclass introduces the EDGeS Grid platform and the EDGeS Application Development Methodology. Case studies of tools that make the application development easier and examples for applications that have been ported to EDGeS using the EADM are presented.

Tamas Kiss is a Senior Lecturer in Database Systems at the Department of Information Systems and Computing, and a researcher at the Centre for Parallel Computing at the School of Informatics, University of Westminster, London. His research interests include parallel and Grid computing, and he has extended experience in the area of legacy code deployment, interoperation of Grid systems, and application porting to service and desktop Grid systems. He led the design and development activities resulting in the Grid Execution Management for Legacy Code Architecture (GEMLCA) solution, now a Globus incubator project, within the UK EPSRC founded OGSA Testbed project. He contributed to the CoreGrid Network of Excellence project as the leader of the Legacy Code Wrapping and Deployment Methodologies Research Group within the Institute on Grid Systems, Tools and Environments. He currently leads the Grid Application Support Service activity within the European EDGeS project.
Tamas has extended experience in teaching in higher education and giving Grid tutorials, lectures and hands-on sessions (e.g. GEMLCA/P-GRADE portal courses organised by the UK National e-Science Center (NESC) and the EGEE project). He co-authored one book and more than 50 scientific papers in journals and conference proceedings, and as book chapters.