EC Expert Group on Next Generation GRIDs

by Keith Jeffery

The third Next Generation Grid expert group (NGG3) convened by the European Commission has completed its work and reported. The report (and much relevant documentation on GRIDs) is available at http://www.cordis.lu/ist/grids/

In the past few years, a group of high-level experts, named the Next Generation Grid (NGG) expert group, has developed a vision that has emerged as the European vision for Grid research. Driven by the need and opportunity of bringing Grid capabilities to business and citizens, the NGG vision underpins the evolution of Grid from a tool to solve compute- and data-intensive problems towards a general-purpose infrastructure enabling complex business processes and workflows across virtual organisations (VOs) spanning multiple administrative domains.

The NGG vision, articulated by NGG1 (2003) consists of three complementary dimensions: the end-user perspective where the simplicity of access to and use of Grid technologies is exemplified; the architectural perspective where the Grid is seen as a large evolutionary system made of billions of interconnected nodes of any type; and the software perspective of a fully programmable and customisable Grid. In order to realise the Next Generation Grid vision, numerous research priorities were identified in terms of properties, facilities, models, tools, etc. which have inspired national and international research programmes for Grid research. Almost half of the NGG1 experts were from ERCIM member organisations. The EC FP6 Call2 in the area of GRIDs resulted in projects aligned with the NGG1 vision, including the Network of Excellence managed by ERCIM: 'CoreGRID'.

NGG2 (2004) went further and elaborated the middleware required for GRIDs and considered the requirements of operating systems to support a GRIDs environment. Particular attention was paid to the need for self-* (self-managing, self-organising, self-healing, self-tuning etc) systems. There was initial consideration of the need for semantic description of service components to permit the construction of such systems. Almost 20% of the NGG2 experts were from ERCIM member organisations: this reflects the broadening acceptance of GRIDs in other organisations represented by the majority of experts. Projects resulting from EC FP6 Call 5 in the area of GRIDs are currently under negotiation.

NGG3 (2005 reporting January 2006) built upon these foundations and concentrated on a service-oriented architecture where the services have strong semantic descriptions allowing self-choreography (composition with flexibility and dynamism). Again approximately 20% of the experts were drawn from ERCIM member organisations and this team had a much stronger participation from industry, indicating the take-up of commercial interest in GRIDs. The SOKU (Service-Oriented Knowledge Utility) vision identifies a flexible, powerful and cost-efficient way of building, operating and evolving IT intensive solutions for use by businesses, science and society. It builds on existing industry practices, trends and emerging technologies and gives the rules and methods for combining them into an ecosystem that promotes collaboration and self-organisation. The benefits are increased agility, lower overhead costs and broader availability of useful services for everybody, shifting the balance of power from traditional ICT (Information and Communi-cation Technology) players towards intermediaries and end-consumers of ICT. It is fortunate that SOKU may also be read as Self-Organising Knowledge Utility and Semantic Oriented Knowledge Utility.

The need for developing the SOKU vision stems from the necessity of effectively bringing knowledge and processing capabilities to everybody, thus underpinning the emergence of a competitive knowledge-based economy. The SOKU vision builds on and extends the Next Generation Grids vision. It captures three key notions:

• Service Oriented - the architecture comprises services which may be instantiated and assembled dynamically, hence the structure, behaviour and location of software is changing at run-time
• Knowledge - SOKU services are knowledge-assisted ('semantic') to facilitate automation and advanced functionality, the knowledge aspect reinforced by the emphasis on delivering high level services to the user
• Utility - A utility is a directly and immediately useable service with established functionality, performance and dependability, illustrating the emphasis on user needs and issues such as trust.

The primary difference between the SOKU vision and earlier approaches is a switch from a prescribed layered view to a multi-dimensional mesh of concepts, applying the same mechanisms along each dimension across the traditional layers.

Thanks to the substantial investments and the numerous initiatives launched at the Member States and European levels, Europe has succeeded in establishing a leading worldwide position in Grids. The consistent portfolio of Sixth Framework Programme (FP6) Grid research projects will further contribute to the realisation of the NGG vision, thus boosting European competitiveness in Grid technologies and applications. It is no accident that ERCIM experts have been involved heavily in this strategic work. Three ERCIM personnel have been involved in each of the 3 expert groups: Thierry Priol (INRIA), Domenico Laforenza (CNR) and the author. We have worked closely and productively with our EC colleagues, particularly Franco Accordino, Max Lemke and Wolfgang Boch.

Link:
http://www.cordis.lu/ist/grids/

Please contact:
Keith G Jeffery, Director, IT CCLR and ERCIM president
E-mail: K.G.Jefferyrl.ac.uk