| The information below relates to the initial phase of the edikt project, which ended in May 2005. Information on the current phase is available via the edikt portal. |
| edikt phase one | ||||
| about | products | activities | ||
|
|
e dikt ::EDG-RLS |
|
|
Background The Large Hadron Collider (LHC) is expected to generate about 5 petabytes of experimental data a year (~500 GB / hour). The DataGrid project (funded by the European Union) has been set up to address the computing requirements for the LHC. The objective of the DataGrid project is to build the next generation of computing infrastructure, which will provide a computation resource to help with the analysis of shared large-scale databases, from hundreds of TeraBytes to PetaBytes, across widely distributed scientific communities. The Data Management work package is part of the DataGrid. The goal of this work package is to develop the middleware infrastructure to coherently manage and share petabyte-scale of information across the grid. The infrastructure will allow substantial amounts of data to be moved and replicated securely, at high speed from one geographical site to another, and to manage synchronisation of remote replicas. Part of this infrastructure is the European Data Grid Replica location service (EDG-RLS). The EDG-RLS which was known as Giggle (GIG-a-scale Global Location Engine) is a system that maintains information about the physical locations of data and provides access to this information. For some databases, it is often desirable to create remote read-only copies of data elements. Given a unique logical identifier, EDG-RLS will determine the physical locations of one or more copies of this content. EDG-RLS will also maintain and provide access to information about the physical locations of the copies. Evaluation of the EDG-RLS The original intention was to install the European Data Grid data replication management system and evaluate its suitability as a replica manager for experimental particle physics. At the time of the evaluation the software was not mature enough to get to any useful conclusions about its suitability. Our primary conclusion is (at the time of the evaluation) that the EDG Replica Management Service software was still too immature to use without detailed insider knowledge. The installation of the Testbed 2.0 version of EDG-RLS was much more straightforward than that for earlier versions, and once installed it passed all the fidelity tests. A further evaluation might be considered when the middleware infrastructure is released at least in beta form. This project is not currently active.
|
||
|
|
||