A vision for European Grid research

Brussels, 16 Sep 2004

With hundreds of technical experts, business leaders and policy makers gathered in Brussels on 15 September to mark the launch of 12 EU funded Grid research projects, the Commission took the opportunity to organise a plenary session to consider a strategic research agenda for this emerging high-tech sector.

The Grid has been defined as an environment enabling coordinated resource sharing and problem solving among individuals, institutions and virtual organisations. Given that the first Grids were developed within different research communities in response to an ever-growing demand for computational power, however, there is as yet no single blueprint for their design and construction, and much remains to be defined.

With billions of euro beginning to pour into Grid research programmes across the globe, though, it appears inevitable that these high-speed networks will be a feature of our technological future. According to the Peter Zangl, Deputy Director of the Commission's DG Information Society, in the coming decades it won't just be researchers using them either: 'Grids have the potential to revolutionise the way business is conducted.'

Mr Zangl believes that Grid technology will deliver five clear benefits: it will improve the quality of products and services, reduce the total cost of ownership, enable the creation and provision of 'anywhere, anytime' services (so-called ambient computing), contribute to the next generation of Internet, and provide a crucial backbone infrastructure.

With such hopes surrounding the technology, the Commission has made Grids one of its highest strategic objectives within the information society technologies (IST) priority of the Sixth Framework Programme (FP6), and has established a Grids unit to oversee the programme's implementation. The 12 projects that the Commission presented to participants, including several large-scale international initiatives, represent a total EU investment of 52 million euro, the first tranche of a 120 million euro total investment in Grids until 2006.

With such levels of funding and organisation for research, complemented by various national initiatives, the Commission expects Europe to gain a leadership role in Grid technology. But much of the research and a great deal of innovating remains to be done, and those with a stake in the process spent the morning session discussing the milestones that would guide their way.

Fujitsu's David Snelling has already given much consideration to the subject, as chair of the Commission's Next Generation Grid Expert Group. The group was put together to plot a course beyond the current scientific Grids with a five to seven year timeframe in order to focus current work in the right areas.

The group considered the Grid of the near future from three perspectives: that of the end user, the architectural perspective, and the software perspective. According to Dr Snelling, from the users' point of view the Grid will be able to carry out millisecond processes using century-long data sets, and they will notice a market 'virtualisation' away from physical mechanisms.

Unlike today's Internet, where people are on the outside looking in, Grid users will themselves be 'nodes' on the network. 'You don't just ask questions of the Grid, it can also ask them of you,' says Dr Snelling. And as we move further down the path leading from data to information, and from information towards knowledge, the Grid will provide a powerful yet affordable knowledge discovery service.

From an architectural perspective the Grid will be virtual and pervasive - always available in our daily lives whenever we need it. Its dynamic service provision essentially means that services will be able to create themselves according to a user's needs, in what Dr Snelling calls a 'self-organising ecology'. In terms of its architectural properties, the Grid must be simple, resilient, straightforward to administrate, and 'transparent', in the sense that a user will not notice all of the network's individual nodes. The major software issue at this stage is interoperability, which requires collaboration on standards. 'There will be no single Grid, so interoperability issues are absolutely key,' argues Dr Snelling.

Having built a picture of the next generation of Grids, the Expert Group has defined a way forward for the EU. The primary aim, it says, is to maintain EU leadership in Grid research, business development and standardisation. In order to do so, it will have to exploit areas of traditional EU strength, such as its social diversity, its collaborative approach to research and business, its track record in semantics and ontology research, and its experience in embedded systems.

Already, Europe's character has imposed itself on their collective vision for the Grid, according to Wolfgang Boch, Head of the Grid Technologies Unit within DG Information Society. The EU's vision includes much more of a societal dimension than, for example, the US, where the emphasis is on supercomputing capabilities. Europe has chosen to place an emphasis on serving the whole community on the full range of technical devices, he said.

The areas where Europe needs to improve its performance are in the communication and exploitation of its research results, argued Mr Boch. 'This is the most significant area for future improvement of efforts. We have an excellent research capacity in Europe, but too little activity in terms of commercialisation,' he added.

Isidoro Padilla, CEO of Spanish-based global telco operator Telefónica, believes that this is because, until now, Europe's Grid community has been too academic, with projects being launched purely from a technical perspective with no thought for business opportunities. 'Grids offer opportunities for new business units inside telcos, but we need more research focused on delivering a return on investments.'

According to Mr Padilla, Grid products with the potential for commercial exploitation could include e-health solutions, such as dynamic healthcare systems, e-learning services, providing huge computational and knowledge resources, and disaster response, delivering intelligent support for decision making. 'All are areas of high value to users, and we could quickly export the results to other commercial areas,' he explained.

All who contributed to the session agreed that a great deal of research and other work is needed before any vision can be given a more solid form, and that significant challenges will present themselves along the way. However, with the resources in place and the collective will to realise their vision apparently secured, it could only be a matter of time until the next generation Grid is with us. In the words of Wolfgang Boch: 'Grids are not just another technology - they will change the way we all work and live.'

For further information, please consult the following web address:
http:///www.cordis.lu/ist/grids/index.htm

CORDIS RTD-NEWS / © European Communities
Item source: http://dbs.cordis.lu/cgi-bin/srchidadb?C ALLER=NHP_EN_NEWS&ACTION=D&SESSION=&RCN= EN_RCN_ID:22626