Article

Building a production grid in Scandanavia

Uppsala University, Uppsala, Uppsala, Sweden
IEEE Internet Computing (Impact Factor: 2). 08/2003; 7(4):27- 35. DOI: 10.1109/MIC.2003.1215657
Source: IEEE Xplore

ABSTRACT Innovative middleware solutions are key to the NorduGrid testbed, which spans academic institutes and supercomputing centers throughout Scandinavia and Finland and provides continuous grid services to its users.

0 Followers
 · 
140 Views
  • Source
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In a grid computing context the execution of jobs on remote machines of collaborating institutes often requires the provisioning of more than mere CPU time. Besides libraries and utilities distributed with the operating system in question, further software may be required by the jobs and in such cases need to be installed prior to the job's execution. In grid computing this problem is aggravated by decreased personal contacts among collaborators. Currently, Virtual Organisations of existing production grids typically agree on a set of runtime environments that participating sites install manually. The challenge is to automate this process, thus easing the burden of the site's maintainers and allowing grid-wide software up- dates with immediate eect. This paper presents an RDF based schema for the description of runtime environments and the implementation of a service for their automated and dynamic installation.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The NorduGrid middleware, ARC, has integrated sup- port for querying and registering to Data Indexing services such as the Globus Replica Catalog and Globus Replica Location Server. This support allows one to use these indexing services for for example brokering during job- submission, automatic registration of files and many other things. This integrated support is complemented by a set of command-line tools for registering to and querying these Data Indexing services. In this article we will describe experiences with these indexing services both from a daily work point of view and in production environments such as the Atlas Data- Challenges 1 and 2. We will describe the advantages of such Data Indexing services as well as their shortcomings. Finally we will present a proposal for an extended Smart Indexing Service which should deal with the shortcomings described. Such an indexing service is being designed at the moment.

Full-text (2 Sources)

Download
35 Downloads
Available from
Jun 2, 2014