Architecture for Integrating Computational Tools Based on Grid Services for System Monitoring and Alerting

Abstract

This article presents examples of computer architecture based on Grid services with the possibility of application to systems of hydrometeorological monitoring and alerting, as well as a support for civil defense. The proposed architecture benefit applications based on rainfall and hydrological simulation forecast. Computer architecture described here integrates tools for ad-hoc system of monitoring and alerting the Itajaí river basin. This system is intended to help in decisionmaking during a crisis situation, since its features are: level prediction, weather forecasting, flood spots, simulation scenarios, with the integration of these tools, including those based GIS and Grid services. These tools can be effective early warning systems also in normal situations, i.e. when there is no risk to any climate events. In this case, a fictional rainy weather event could be simulated and their impact studied. One of the main benefits of the proposed architecture is related to the total execution time for applications with high computational cost. Intuition is then that the computer architecture based on Grid computing, consisting of several virtual organizations which provides shared resources, that can meet the main demands of a system of monitoring and early warning for floods, more specifically for the warning system CEOPS / FURB as well as the details of the demanded information by civil defense.

Cloud Computing Architecture for Monitoring and Flood Warning System: Conceptual Aspects

“…System Requirements The integration of tools and data in a proper runtime are very important for decision making before the occurrence of an event characteristics. However, the architectural model for alert systems must offer the following essential requirements [6] and [7]: 9 Response time -for decision making, the results of the forecasts and hydrologic simulation must be submitted in the shortest possible time, thus the ability of computational processing is crucial to meet this demand; 9 Availability -if a service changes its location, condition, or in the presence of a hardware failure, the system should be able to manage this situation, so that the request is granted, the completed task, and the system will continue functioning normally transparently the user; 9 Reserve resources -it is desirable that a model for monitoring and alerting meets some criteria of quality of service. By booking and request on-demand resources can ensure the accomplishment of the task; 9 Real Time -the data must be collected in real time and the information should be generated by CEOPS before the occurrence of the event and with enough time for the Civil Defense agencies may implement the plan for defense.…”

“…Based on this, this requirement is to ensure that the system will meet the processing of certain tasks in certain dimensions of time; 9 Flexibility -is the ability of the system to integrate different tools applied to monitoring and flood warning in the watershed. For example, tools based on numerical modeling and simulation to conduct hydrological and meteorological forecasting, among others; 9 Scalability -applications based on models that require large computational processing capacity, usually have a high degree of parallelization of code [6], so the system must be able to scale from a few to thousands of computing resources.…”