Tutorials at OGF-Europe's 2nd International Event

Tutorials at the 4th EGEE User Forum/OGF25 & OGF-Europe's 2nd International Even, 2-6 March 2009, Catania, Italy

OGF-Europe is organising a series of tutorials at its 2nd International Event, with the aim of  transferring knowledge for the implementation of OGF specifications as a key enabler for the business and scientific communities. Tutorials focus on the management of computational activities; effectively accessing a grid infrastructure and how to make a grid e-infrastructure sustainable.

Managing Computational Activities on the Grid - from Specifications to Implementation Speakers: Balazs Konia, Morris Riedel, Aleksandr Konstantinov, Augusto Ciuffoletti

Monday 2 March 2009, 9:00-10:00

Focus & Target Audience
The tutorial opens with an overview of OGF standards that govern the job submission process together with an introduction to Glue and DEISA's approach to the implementation of security aspects that are vital in a production environment. Demos with real-scaleimplementations are presented to show the job submission interface provided by ARC, NorduGrid and Unicore. 
 
The Tutorial is targeted at project managers and designers. Partcipants should possess a general knowledge of grid computing concepts and superficial knowledge of the XML language. The Tutorial will demonstrate the valuable role played by standards in the design of a job submission interface and the degree of freedom left to a specific implementation.
 
Background
The Interoperability of grid platforms is only possibly through the implementation of accepted standards. The interface for job management is a basic blocking of a grid infrastructure with the specific interface used plays a key role in the successful provision of this service.
 
OGF's Job Submission Description Language, JSDL, describes the resource requirements and other basic characteristics of a job, using an XML document that follows a standard schema. This basic schema is introduced along with the extensions relevant to High Performance Computing (HPC). The Basic Execution Service (BES) describes the dynamic features of a job, from its submission to the time when it leaves the system. Standards provide a basic guideline to ensure interoperation betwen distinct implementations. Glue is a model that enables a fine-grain description of resources for use within the JSDL and BES standards.
 

How new communities can get access to a Grid infrastructure
Speakers: Oxana Smirnova, David Fergusson, Morris Rieder

Monday 2 March 2009, 11:00-12:30

Focus & Target Audience

A community of users may see the services offered by a grid infrastructure as a favourable entry point in activities that require dvanced processing resources: computational power and storagecan be acquired without investing in logistics and training. Several European projects have pionereed the field of providing grid services mainly to scientific communities which need leverage large-scale processing resources without engaging in its deployment.
 
This tutorial brings together three major EU providers to focus on how new communitities can access a grid infrastructure. Participants will learn how these infreastructures meet the needs of a community willing to use the computational services they provide and the role in OGF standards in simplifying training and use of a platform to enable the portability of grid applications. Decision-makers will come away with a clear idea of the advantages to be gained from using a grid infrastructure that complies with OGF standards.
 
Background
NorduGrid is a Grid Research and Development collaboration aimed at developing, maintaining, and support of open source grid middleware, known as the Advanced Resource Connector (ARC). The Nordic DataGrid Facility (NDGF) is the largest production infrastructure that relies on ARC and leverages existing national computational resources and grid infrastructures.
EGEE (Enabling Grids for E-sciencE) is the largest multi-disciplinary grid infrastructure in the world, which brings together more than 140 institutions to produce a reliable and scalable computing resource available to the European and global research community. At present, it consists of approximately 300 sites in 50 countries and gives its 10,000 users access to 80,000 CPU cores around-the-clock.
The DEISA supercomputing Grid is a European research infrastructure resulting from the integration of national High Performance Computing (HPC) infrastructures. This integration of national resource using modern grid technologies is expected to contribute to a significant enhancement of HPC capability and capacity in Europe.

 

How to make a Grid e-Infrastructure sustainable
Speaker: Pasquale Pagano

Wednesday 4 March, 16:00-17:30

Focus & Target Audience

This tutorial, which targets decision-makers and site managers, presents the gCube approach towards a sustaionable grid e-Infrastructure. The main focus is on providing an overview of the benefits in terms of deployment costs, time-frames, and manpower requirements, as well as indsight into how organisations can leverage grids. The tutorial will demonstrate how gCube eliminates manual deployment overheads, ensures optimal placement of services within the infrastructure and opens up important opportunities for outsourcing state-of-the-art implementations to grid-enabled e-Infrastructures.

Participants will gain importantant awareness about issues connected with autonomic grid service deployment in distributed e-Infrastructure and how the adoption of standards contributes to their solution.

Background

By definition, an e-Infrastructure is a framework enabling secure, cost-effective and on-demand resource sharing across organisational boundaries. In this context, "resource" indicates a general entity, physical storage and computing resources or digital (software, processes, data), which can be shared and made to interact with other resources to synergistically provide functions serving its clients, either human or inanimate. Thus, an e-Infrastructure poses as a ``broker'' in  a market of resources with the role of accommodating the needs of resource providers and consumers. The infrastructure layer gives support to: 1) resource providers, in ``selling'' their resources through it; 2) resource consumers, in ``buying'' and orchestrating such resources to build their applications. Further, it provides organisations with logistic and technical aids for application building, maintenance, and monitoring.

A well-known instance of such an e-Infrastructure is represented by the grid where a service-based paradigm is adopted to share and reuse low-level physical resources. Application-specific e-Infrastructures are in their turn inspired by the generic e-Infrastructure framework and bring this vision into specific application domains by enriching the infrastructural resource model with specific service resources, i.e. software units that deliver functionality or content by exploiting available physical resources.

This is potentially not-limited market of resources allows a new development paradigm based on the notion of Virtual Research environment (VRE), a.k.a. Collaboratory. This is built by aggregating the needed constituents after hiring them through the e-Infrastructure. In this development paradigm, VREs are considered as organised views built atop the pool of available assets, ranging from computers and servers to collections and services, and the VRE management enabling technology is the one that takes care of the definition and operation of such views. To make this novel paradigm working three fundamental facilities are needed: 1) a mechanism operating the e-Infrastructure while guaranteeing all the involved parties about the Quality of Service; 2) a mechanism supporting VO communities in easily characterizing the VRE they are interested in; 3) a mechanism guaranteeing the deployment and operation of the defined VREs, in addition to a comprehensive pool of resources.