Peter Brezany, Ivan Janciak, Alexander Woehrer and Ibrahim Elsayed
Building Intelligent Grid Systems
Built on the Internet and the World Wide Web, the
Grid is becoming a crucial technology for science
and economy. Since its early years the Grid technology
experienced several development phases or generations.
The latest developments focus on intelligent Grid
systems based on the Semantic Web, Artificial Intelligence, and
advanced software technologies.
This tutorial presents the framework called GridMiner developed within a joint project of the University of
Vienna and the Vienna University of Technology. First, the Gridminer architecture will be presented.
It includes advanced interactive knowledge discovery workflow management, GUI, services for data mediation, data preprocesing
data mining, and OLAP. The GridMiner functionality will be presented by a live demo. The future developments of
the GridMiner system will be also discussed.
Second, the methodology for application of GridMiner to real
data discovery tasks will be explained.
Presentation handouts will be available for the tutorial participants.
More information about the GridMiner framework can be found at
Piotr Siwczak, Pawel Wolniewicz, Marcin Radecki, Tomasz Szepieniec
EGEE gLite and CrossGrid Performance Tools
The tutorial is designed for new and prospective users of EGEE infrastructure who want to get familiar with Grid Computing. It covers introduction to basic grid concepts, grid middleware components and gaining an end-user experience from interacting with the Grid. The focus is aimed at the upcoming version of EGEE middleware - gLite.
In the first part an overview of the EGEE infrastructure is given. It highlights potential benefits one can derive from using the Grid and describes briefly the computational model provided by EGEE.
The second part focuses on the Grid middleware. It describes its components and presents how they interact with one another during job processing. Resource and data management facilities are covered here. This section of the tutorial is based on the new grid middleware - the successor of LCG called gLite - highlighting major differences/improvements from LCG middleware.
The third part is a hands-on guide with practical examples on how to submit and then control jobs on the Grid and how to provide data control using gLite. Additionally, OCM-G and G-PM – tools for application performance monitoring in grid environment developed in CrossGrid Project will be presented.
e-Science and Grid - The Virtual Lab for e-Science Approach
A short overview will be presented of the challenges and research issues in Grid and e-Science. It will be outlined what the drivers for Grid as well as e-Science developments are
and what consequences this has for their respective developments. The basic ideas behind our Virtual Laboratory project will
be presented, as well as the research philosophy applied. Illustration will be chosen from the life science domain.