Introduction – General Description and Milestones. The context Many Civil Protection applications require a strict integration with research infrastructure providing resources (computing, data, services, knowledge, expertise, etc.) useful in the full cycle of emergency management (forecasting, warning, management, assessment). Moreover these activities typically involve many different actors (Civil Protection systems, public bodies, research centers, etc.) that need to share resources in a coordinated and effective way. The recent development in the information and communication science and technology, provide the basis for enabling such applications. In particular, Grid technologies are designed just to allow the coordinated sharing of resources (computing, storage, communication, sensors) needed by such Virtual Organizations (VOs). Consequently, the adoption of a Grid-based infrastructure seems a natural choice to build a cooperative platform for supporting Civil Protection activities. However Civil Protection applications have specific requirements such as: - especially during emergency situations, they require to access research infrastructure to run models or search information, but they need real- time (RT) or near-real-time (NRT) responses, privileging time of response instead of accuracy. - during emergency situations, they gain great benefit from the possibility to control sensors networks and acquisition systems to modify their acquisition strategy and processing chain; - they require to share geo-spatial information that has specific characteristics. In particular, remote-sensing observation from new satellite sensors produces huge amount of data that are frequently updated; - Civil Protection systems often need to interact with military resources so they have the strict data policy and security requirements typical of dual systems (civil/military); - they require knowledge-based systems that starting from data, can provide information suitable by the decision-makers; These requirements have strong impact on research strategies (new data models, algorithms, methodologies need to be investigated) and consequently on the functional and non-functional requirements of the enabling research infrastructures (new protocols, services and architectures are to be designed and implemented). Current Grid platforms were mainly designed for researches requiring intensive processing and data management, thus they offer high computing and storage capabilities. Only in recent times Grid research is facing new pr...
Introduction – General Description and Milestones. This section describes in detail the JPA, which is planned for months 25-42. The aim for the first eighteen months is summarised as “setting the stage” and it is the phase of getting to know each other, making inventories of current practice and research, establishing the communication platform, start sharing current technologies and procedures, and laying the foundation for joint research, including by initiating shared PhD projects. The major deliverables for the first phase will be the publication of calls for new partners, identification of funding possibilities for joint research (e.g. other FP6 calls, various national and regional opportunities), harmonising current practice and protocols, laying the foundations for shared facilities and databanks systems, and establishing joint training programmes. Also, the dissemination concepts will be established throughout Europe, and communication with key stake holders in the relevant areas will be set up. These generalities have been translated into 18 dedicated WPs, organised as the first major blocks of work of the corresponding activities. They are grouped in four horizontal platforms of WPs. The eight Integrating WPs (IA1.1-1.8) address Strategic Objective 1 (To identify, address and overcome technical changes) and aim at harmonising, stimulating and facilitating new technology, informatics and systems for common use. These will form the basis for the four Joint Research WPs (RA2.1-2.4), which address Strategic Objectives 2 & 3 (To identify and provide new information for missing data and foods, and to identify user and stakeholder requirements) aim at exploiting the technological and scientific developments relating to databank infrastructure and specifications in order to enhance the quality of food databank linking, coverage and management. The Spreading of Excellence WPs (SA3.1-3.5) address Strategic Objectives 4 & 5 (To spread excellence and enhance impact, and to identify socio-economic and sustainability impacts] build upon the acquired knowledge to share this with target user and stakeholder groups (researchers, industry, society, healthcare), and to establish long-term durability for the network. Lastly, the Management Workpackage (MA1) describe the co- ordinated activities to flexibly structure the Network, achieve and monitor integration, and the procedures for SME enrolment and participation. Managing risks and identifying contingency plans for these The assessment of the progress of EuroFIR ...
Introduction – General Description and Milestones. Multi-tier applications are the mainstay of industrial applications. A typical example is a two-tier application, consisting of a client talking to a server, which itself interfaces with a database (see top of the figure). The business logic is executed at the server and the application data and metadata are stored on the database. But multi-tier applications are brittle: they break when exposed to stresses such as failures, heavy loading (the “slashdot effect”), network congestion, and changes in their computing environment. In practice, these applications require intensive care by human managers to provide acceptable levels of service. This becomes especially cumbersome for large-scale systems. For example, deploying a distributed file system across several organizations requires much manual configuration, as does adding another file server to the existing infrastructure. If a file server crashes, most file systems will stop functioning or fail to provide full service. Instead, the system should reconfigure itself to use another file server. This desirable behavior is an example of self management. In the SELFMAN project, we will show how to make large-scale distributed applications self managing. We will implement a general architecture for these applications by combining research on structured overlay networks together with research on component models. These two areas each provide what the other lacks: structured overlay networks provide a robust communications infrastructure and low-level self-management properties, and component models provide the primitives needed to support dynamic configuration and enable high-level self-management properties. To show the effectiveness of this architecture, we will develop a J2EE demonstrator, where standard J2EE functions are deployed over a large scale network and are extended with self-management capabilities (see bottom part of the figure). We will evaluate the usefulness of the self-management abilities both quantitatively and qualitatively using industry data. The foundation of SELFMAN will be a combination of a structured overlay network with a component model. Both areas have much matured in recent years, but they have been studied in isolation. It is a basic premise of SELFMAN that their combination will enable achieving self-management in large-scale distributed systems. This is first of all because structured overlay networks already have many low-level self-management properties. Structured overlay ne...
