The current production software platform for EGEE is LCG-2 which can be to connect your site to the grid (the installation requires system manager privileges).
Here you can find an a pdf overview of how to use this platform as a user.
gLite is currently using a number of GT2 services. GT4 has been released recently and JRA1 is currently evaluating it. Depending on the evaluation, and compatibility other external dependencies with GT4 a roadmap to migrate from GT2 to GT4 will be established. Some GT4 components such as the WSS are already being prototyped. Globus is a full partner of JRA1.
Unfortunately gLite doesn't yet support 64 bit architectures, however we are currently in the process of porting to IA64. Neither gLite 1.5.0 nor the release integrated with LCG-2, gLite 3.0.0, that we are preparing now, are ready for 64-bit architecture as yet.
Any version of Scientific Linux, in particular the 3.0.x series, can be used. There is no experience yet with Scientific Linux 4.0. Further questions on specific problems may also be directed to the mailing list.
There are a couple of guides available for the introduction of the use of the EGEE Middleware, gLite. gLite is being developed, and is based on the LCG stack. In gLite version 3.0.0, scheduled for early 2006, the convergence between the two software stacks will be complete and all LCG Middleware references will be replaced with gLite 3.0.0.
There is a documentation section being currently developed on this website, but you can find a great deal of information at the following links;
• Simplified access - EGEE will reduce the overhead of separate accounting systems by providing means for users to join virtual organisations with access to a Grid containing all the operational resources the user needs.
The grid operations groups in the EGEE project will ensure the delivery of production level Grid services, the essential elements of which are manageability, robustness, resilience to failure, and a consistent security model, as well as the scalability needed to rapidly absorb new resources as these become available.
A Grid Middleware Engineering and Integration Activity will support and continuously upgrade a suite of software tools capable of providing production level Grid services to a base of users which is anticipated to rapidly grow and diversify.
The middleware activities in EGEE focus primarily on re-engineering existing middleware functionality, leveraging the considerable experience of the partners with the current generation of middleware. The middleware will evolve towards a service oriented architecture (SOA) based on standards developed within the Web Services and related communities. Middleware Re-engineering Centres will take responsibility for the following key services:
Quality Assurance and Grid Security teams will be closely connected to this effort. A middleware Integration and Testing Centre will be based at CERN. As a starting point for this activity, an effort has been launched to integrate some of the leading Grid middleware from the Virtual Data Toolkit (VDT), The European DataGrid Project (EDG) and the AliEN project.
Porting of gLite middleware to other platforms is planned for the future but we do not yet know the timeline for porting to Windows.
For the time being PVM is not supported. The only supported way to run parallel jobs is through MPI. An extension to support to PVM, although possible, it is not planned at this time.
Standard packages. A complete list can be found from:
This is the next generation middleware in EGEE. A mixed LCG/gLite middleware distribution is currently being tested and will be deployed on the production infrastructure in April 2006.
Myrinet is a network connection between the worker nodes in the computing cluster at this site. Myrinet is typically used by parallel jobs that communicate between processes (typically using MPI) on different compute nodes. So: will it be used? Yes, if the application that runs on this cluster is a parallel application. If the application that is run is a simple one job per node then: no, myrinet will not be used. Of course, in the first case, the cluster must be set up to run MPI jobs, and must publish this appropriately in the information system so that MPI jobs can be scheduled to run there.
2nd EchoGRID Strategic Workshop, 29-30 October 2007, Beijing, China: Towards a Shared EU & Chinese Vision for Grid Research Perspectives... Read more
Try the GRID
Become a User
Want to become a user of the EGEE Grid? Click here
If your project is related to EGEE, please register it
Click here for all the EGEE Information Sheets.
Get in touch with the EGEE Project Office at .
If you have questions about the Website contact: .
EGEE in Other Countries: | | | | | | | | | | | | | | | | | | |
EGEE-II is a project funded by the European Union under contract number INFSO-RI-031688