Earth observation (EO) products provided by different EO facilities differ in terms of format, geographic reference system, projection and so on. Usually, domain spe- cialists are increasingly willing to integrate, compare and fuse data coming from different EO sensors. EO systems are thus characterized by the management of het- erogeneous data. Moreover, several processing algorithms are composed by a lot of steps and some intermediate partial results are used as input to other algorithms defining a dataflow graph. Grid Computing Environments represent a valid support to handle the data heterogeneity and the complexity of the entire processing provid- ing the user with powerful tools for resource orchestration and data management. A Problem Solving Environment (PSE) based on grid technologies can be thought as the user side of a grid computing system where users interact via user-friendly grid portals or desktop applications which in turn handle a set of distributed back-end resources. From this point of view the workflow represent a convenient approach that allows users to compose complex services starting from single services and/or standalone applications and specifying the data flow among the involved applica- tions and a set of constraints. In this paper we present the design and development of a grid problem solving problem environment aimed at EOS applications.
GridFlow for Earth Observation Data Processing
ALOISIO, Giovanni;CAFARO, Massimo;EPICOCO, Italo;
2005-01-01
Abstract
Earth observation (EO) products provided by different EO facilities differ in terms of format, geographic reference system, projection and so on. Usually, domain spe- cialists are increasingly willing to integrate, compare and fuse data coming from different EO sensors. EO systems are thus characterized by the management of het- erogeneous data. Moreover, several processing algorithms are composed by a lot of steps and some intermediate partial results are used as input to other algorithms defining a dataflow graph. Grid Computing Environments represent a valid support to handle the data heterogeneity and the complexity of the entire processing provid- ing the user with powerful tools for resource orchestration and data management. A Problem Solving Environment (PSE) based on grid technologies can be thought as the user side of a grid computing system where users interact via user-friendly grid portals or desktop applications which in turn handle a set of distributed back-end resources. From this point of view the workflow represent a convenient approach that allows users to compose complex services starting from single services and/or standalone applications and specifying the data flow among the involved applica- tions and a set of constraints. In this paper we present the design and development of a grid problem solving problem environment aimed at EOS applications.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
