Biocomputing Unit / Grid computing

RESEARCH

Grid-oriented computing

Introduction

Advances in networking technology make it possible to construct large scale high-performance distributed computing environments that integrate, through high-speed networks, complex heterogeneous resources, such as supercomputers, large databases, high performance storing devices, advanced visualization devices and/other scientific instruments with the aim to create networked virtual supercomputers or metasystems, coming into reality the concept of metacomputing(Catlett and Smarr, 1992; Grimshaw et al., 1998). Such an infraestructure is known as Computational Grid (Foster and Kesselman, 1998).

Computational grids have much in common with both distributed and parallel systems, yet also differ from these two architectures in important ways. Like a distributed system, a networked supercomputer must integrate resources of widely varying capabilities, connected by potentially unreliable networks and often located in different administrative domains. However, the need for high performance can require programming models and interfaces radically different from those used in distributed systems (Foster and Kesselman, 1997). As in parallel computing, metacomputing applications often need to schedule communications carefully to meet performance requirements. However, the heterogeneous and dynamic nature of metacomputing systems limits the applicability of current parallel computing tools and techniques (Foster and Kesselman, 1997).

Roughly, computational grids are characterized by scalability, heterogeneity at multiple levels, unpredictable structure, dynamic and unpredictable behaviour and, finally, multiple administrative domains. Fundamental to all of these issues (Foster and Kesselman, 1997) is the need for mechanisms that allow applications:

to obtain real-time information about system structure and state,
to use that information to make configuration decisions, and
to be notified when information changes.

Clearly, the success of computational grids will depend on the existence of grid-specific middleware (Foster and Kesselman, 1997; Foster and Kesselman, 1999) that addresses the need of computations, including dynamic resource allocation, communications management, system information services, process creation, remote data access, etc. The CNB is involved in several national and international projects using the Grid computing as link.

Grid Initiatives

This is one of the most important initiatives/projects in the international forum using Grid technology. Managed and supervised by CERN (European Organization for Nuclear Research) the main goal is to connect the biggest number of institutions at Europe and the rest of the world (27 total). [ More info ].

The CNB is collaborating in this project, involved in the NA4 Biomedical software. Moreover, it shares its resources (a cluster of 8 biprocessor machines, 2.0 GHz per processor), as the rest of the EGEE partners do.

If you need more info about the particular applications that we have migrated using this technology, see the EGEE biomedical applications list

IRISGRID is a spanish initiative (using the EGEE GRID middleware) for making a spanish GRID between the most important research institutions. [ More info ].

References

C. Catlett and L. Smarr. (1992) Metacomputing. Communications of the ACM, 35, 44-52.
I. Foster, C. Kesselman. (1997) Globus: A Metacomputing Infrastructure Toolkit. Intl J. Supercomputer Applications, 11, 115-128.
The Grid: Blueprint for a New Computing Infrastructure. (1998) Edited by Ian Foster and Carl Kesselman Morgan Kaufmann Publishers. 1998.
A. Grimshaw, A. Ferrari, G. Lindahl, K. Holcomb. (1998) Metasystems. Communications of the ACM, 41, 46-55.
I. Foster, C. Kesselman. (1999) The Globus project: a Status Report. Future Generation Computer Systems, 15, 607-621.
I. Foster (2000). Internet Computing and the Emerging Grid Nature.
I. Foster, C. Kesselman, S. Tuecke (2001). The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Intl. J. Supercomputer Applications, 15(3), 2001.