Viet Tran

A roadmap for a dedicated Earth Science Grid platform

Due to its intensive data processing and highly distributed organization, the multidisciplinary Earth Science applications community is uniquely positioned for the uptake and exploitation of Grid technologies. Currently Enabling Grids for E-sciencE, and other large Grid infrastructures are already deployed and capable of operational services. So far however, the adoption and exploitation of Grid technology throughout the Earth Science community has been slower than expected. The Dissemination and Exploitation of GRids in Earth sciencE project, proposed by the European Commission to assist and accelerate this process in a number of different ways, had between its main goals the creation of a roadmap towards Earth Science Grid platform. This paper presents the resulting roadmap.

Problem‐solving environment for flood forecasting

This paper presents a prototype of the collaborative problem solving environment (PSE) for flood forecasting. Flood forecasting is a complex problem that requires cooperation of many scientists in different areas. To enable this cooperation in a manner comfortable to hydrometeorological experts, a part of the CrossGrid project is aimed towards developing a PSE, whose prototype is described here. The PSE consists of a cascade of simulation models, a storage system for computed and measured data and other used datasets, a Web‐based portal with collaboration tools and a powerful computation facility. The whole system is tied together by grid technology and is used to support a virtual organization of experts, developers and users.

Flood Forecasting in CrossGrid Project

This paper presents a prototype of flood forecasting system based on Grid technologies. The system consists of workflow system for executing simulation cascade of meteorological, hydrological and hydraulic models, data management system for storing and accessing different computed and measured data, and web portals as user interfaces. The whole system is tied together by Grid technology and is used to support a virtual organization of experts, developers and users.

Hydro-meteorological Scenarios Using Advanced Data Mining and Integration

We present a hydro-meteorological scenario and its adaptation to a novel, advanced data integration and data mining platform developed within the project ADMIRE. The scenario uses data mining techniques instead of more common physical models in order to predict phenomena which are not being ordinarily solved in Slovakia – water temperature and discharge wave propagation downstream of a major water reservoir. The scenario is one of a set of use cases, which form the Flood Forecasting Simulation Cascade – a pilot application of ADMIRE. We describe the data integration methodology approach we have devised, and the variables used in data mining training of the scenario.

Parallel Flood Modeling Systems

Flood modeling is a complex problem that requires cooperation of many scientists in different areas. In this paper, the architecture and results of the ANFAS (Data Fusion for Flood Analysis and Decision Support) project is presented. This paper also focuses on the parallel numerical solutions of flood modeling module that are the most computational-intensive part of the whole ANFAS architecture.

Parallel Program Model for Distributed Systems

In this paper, we present a new model for parallel program development called Data Driven Graph (DDG). DDG integrates scheduling to parallel program development for cluster of workstations with PVM/MPI communication. DDG API library allows users to write efficient, robust parallel programs with minimal difficulty. Our experiments demonstrate the new parallel program model with real applications.

A novel approach for developing interoperable services in cloud environment

Cloud computing has seen a tremendous growth in the last five years. Along with the growth, many cloud models have been marketed. They deliver hardware and software as virtualization enabled services to users. Although cloud computing offers considerable advantages such as unlimited resources, manageability and lower investment costs but there are still barriers to exploit it. One of the barriers is the difficulties, which users are being faced when developing and deploying their own services into clouds. In this paper, we present a novel approach for developing interoperable services that can be deployed in different cloud infrastructures at the same time. The approach provides an instrument with emphasis on abstraction, inheritance and code reuse. Using the approach, cloud-based services are developed easily by extending existing abstractions classes provided by the instrument or other developers. The interoperability between different clouds is solved via the basic abstraction class of the instrument and all services are inherited and benefited from this advantage.

Application of ADMIRE data mining and integration technologies in environmental scenarios

In this paper we present our work on the engine for integration of environmental data. We present a suite of selected environmental scenarios, which are integrated into a novel data mining and integration environment, being developed in the project ADMIRE . The scenarios have been chosen for their suitability for data mining by environmental experts. They deal with meteorological and hydrological problems, and apply the chosen solutions to pilot areas within Slovakia. The main challenge is that the environmental data required by scenarios are maintained and provided by different organizations and are often in different formats. We present our approach to the specification and execution of data integration tasks, which deals with the distributed nature and heterogeneity of required data resources.

ABSTRACTION LAYER FOR DEVELOPMENT AND DEPLOYMENT OF CLOUD SERVICES

In this paper, we will present an abstraction layer for cloud computing, which intends to simplify the manipulation with virtual machines in clouds for easy and controlled development and deployment of cloud services. It also ensures interoperability between different cloud infrastructures and allows developers to create cloud appliances easily via inheritance mechanisms.

Data Driven Graph: A Parallel Program Model for Scheduling

In this paper, we present a new powerful method for parallel program representation called Data Driven Graph (DDG). DDG takes all advantages of classical Directed Acyclic Graph (DAG) and adds much more: simple definition, flexibility and ability to represent loops and dynamically created tasks. With DDG, scheduling becomes an efficient tool for increasing performance of parallel systems. DDG is not only a parallel program model, it also initiates a new parallel programming style, allows programmers to write a parallel program with minimal difficulty. We also present our parallel program development tool with support for DDG and scheduling.