GridCoord DoW

Key Projects

IRISGrid

IRISGrid is a Spanish Grid initiative fostered by several of the most important research institutions with a high interest on Grid computing. IRISGrid aims at providing interested parties with best practice advices on the use of Internet resources and on security issues (i.e., in general, building a good infrastructure to deploy Grid-based applications) and also at achieving higher scientific and technological coordination. IRISGrid was born within RedIRIS, the Spanish national organization which manages the academic Internet network. IRISGrid is a meeting point of many institutions (listed below) and it also maintains contacts with the Thematic Network for Middleware. Besides academic partners, GMV, Telefónica I+D, GridSystems, Interquanta, and Atos Origin are industrial partners of IRISGrid.

The main objectives are:

• 
  Surveying the experience and interests of the Scientific Community.
  
• 
  Proposing a Network of e-Science centres.
  
• 
  Setting up a suitable structure to apply for funds from different sources. The proposed structure is depicted in Figure 3, which also gives an idea of the sources for funding and support available for Spanish institutions, both Spanish and international.
  

The members of the IRISGrid initiative are listed in the table below:

Institution	Research group (if applicable)
RedIRIS
Astrophysics Institute of the Canary Islands (IAC)
Center for Energy, Environment, and Technological Research (CIEMAT)
National Institute for Aerospace Technology (INTA)
Catalonian Supercomputing Centre (CESCA)
Galician Center for Supercomputing (CESGA)
Scientific Information Point of Barcelona (PIC)
Institute of High Energy Physics (IFAE)
Institute for Particle Physics (IFIC)
University of Murcia (UM)
University of Oviedo (UNIOVI)
University of Valencia (UV)
Spanish National Research Council (CSIC)	Physics Institute of Cantabria (IFCA)
	Institute for Particle Physics, Valencia
	Astrophysics Institute of Andalusia (IAA)
CSIC-INTA	Center for Microbiology (CAB)
CSIC & University of the Balearic Islands (UIB)	Mediterranean Institute for Advanced Studies (IMEDEA)
CSIC & University of Salamanca (USAL)	Cancer Research Institute (CIC)
Universidad Complutense of Madrid (UCM)	Group for Distributed Systems Architecture and Security
	Group of Computer Architecture and Technology
Autonomous University of Barcelona (UAB)	Group of Computer Architecture and Operating Systems
Technical University of Valencia (UPV)	Networked and High Performance Computing Group
National Biotechnology Center (CNB)	Biological Computing Group
Technical University of Catalonia (UPC)	CEPBA-IBM Research Institute (CIRI)
Carlos III University of Madrid (UCIII)	Computer Architecture, Communications and Systems group
University of Cantabria (UNICAN)	Computer Architecture Group
	Group of Artificial Intelligence in Meteorology (AIMET)
Technical University of Madrid (UPM)	Laboratory for Computational Fluid Mechanics
University of Granada (UGR)	Information Processing Circuits and Systems Group
University of A Coruña	Computer Architecture Group (GAC)
University of Santiago de Compostela	Computer Architecture Group (GAC)
	High Energy Experimental Physics (GAES)
Autonomous University of Madrid (UAM)	High Energy Experimental Physics
University of Almeria (UAL)	Supercomputing Group
University of Málaga (UMA)	Computer Architecture Section

CABGrid: A Virtual Laboratory for Computational Astrobiology

CABGrid is a project of the Advanced Computing Laboratory at the Centre for Astrobiology (CAB), which tries to bring Grid technology to the Computational Astrobiology field. The aim of this project is to design, set up, and deploy a virtual laboratory for Computational Astrobiology (CAB), called CABGrid, using the technologies provided by Globus as a starting point.

CABGrid will use the Astrobiology as application domain, and more concretely:

• 
  Astronomy: search of extra solar planets.
  
• 
  Astrophysics: nebular coagulation.
  
• 
  Biophysics: self-organization phenomena.
  
• 
  Geology: remote detection, earthquake simulations, planetary atmosphere simulations.
  
• 
  Biology: fundamentals of life, extremophiles, molecular evolution, bioinformatics.
  
• 
  Chemistry: prebiotic chemistry, origin of chirality.
  
• 
  Robotics: remote operation, data acquisition.
  

It will include the following functionalities:

• 
  Creation of a data Grid for the data sets obtained from the experiments (data acquisition) or simulations (data derivation).
  
• 
  Creation of a computational Grid for high throughput datamining, extraction, and analysis of those data sets using data-flow graphs.
  
• 
  Creation of a collaboration Grid to foster human interaction and collaboration.
  
• 
  Creation of a Grid portal to ease and, therefore, increase the utilization of the provided facilities.
  

The participants in this project are the Centre of Astrobiology and the National Institute for Aerospatial Technology (INTA). The project is being funded by research grants from these organizations.

GRIDWAY Framework Project

GridWay is a Globus submission framework that allows an easier and more efficient execution of jobs on dynamic Grid environments. GridWay automatically performs all job scheduling steps, provides fault recovery mechanisms, and adapts job scheduling and execution to changing Grid conditions.

While Grids offer a dramatic increase in the number of available processing and storage resources that are available for applications, efficient job submission and management continue being far from accessible to ordinary scientists and engineers due to their dynamic and complex nature. The aim of the GridWay project is to research and develop the technology required to automatically perform all the submission steps and also to provide the runtime mechanisms needed to dynamically adapt the application execution. The GridWay framework has been devised to reduce the gap between Grid middleware and application developers. GridWay incorporates the following techniques to deal with the dynamicity of the Grid:

Adaptive Scheduling: Given the dynamic characteristics of Grid environments, it is necessary to periodically re-evaluate the initial scheduling. So, the schedule can be dynamically adapted to the available resources and their characteristics, normally considering the number of pending and running jobs, and the history profile of completed jobs.

Adaptive Execution: Adaptive execution can improve application performance by adapting it to the dynamic availability, capacity, and cost of Grid resources. In this case the overhead incurred on by job migration is the key issue that must be considered. The released version does not provide support for self-adaptive applications. An application cannot take decisions about resource selection as its execution evolves, and cannot provide its own performance activity to detect performance slowdown.

The GridWay Project is developed with the participation of Complutense University of Madrid (UCM), the National Institute for Aerospace Technology (INTA), and the Centre for Astrobiology (CAB). The founding source are Spanish Research Grants TIC 2002-00334 and 2003-01321