Table 1 - uploaded by Dana Petcu
Content may be subject to copyright.
Source publication
One reason for using Grid computing for satellite image processing is that the required computing performance or data sharing is usually not available at the user side. Two related scenarios are analyzed in the paper: image processing at the data site using standard tools and remote access to parallel computing facilities. Two Grid services were de...
Contexts in source publication
Context 1
... currently implemented client codes are de- scribed in Table 1. They refer to the remote image pro- cessing facilities. ...
Context 2
... that for a typical satellite image of size 9000 x 8000 pixels more than 8 minutes Figure 2: A first application: detect the changes in a river bed -two images taken at eight year distance are compared; top image -a pseudo-color image using three bands; middle -same sector after eight years; bottom image -river bed and differences between the two images detected using a script-Fu for GIMP (changes are marked with specific colors) Figure 3: A second application: image classification performed on an pseudo-color image resulting by the combination of band 3 in red and band 4 in green -the decision tree described in Gluas [11] and the resulting image A Web interface was designed to help the user to interact with the GIMP Grid service. It exposes the facilities presented in Table 1. Figure 4 presents the GUI. ...
Citations
... Grid scheduling, that is, the allocation of distributed computational resources to user applications, is one of the most challenging and complex task in Grid computing. Nowadays, several are the real-life applications in which Grids are involved; some practical fields are protein folding [9], weather modeling [12], and satellite image processing [11]. One of the most known framework for Grid scheduling is the one introduced by Ranganathan and Foster in [13]. ...
Grid scheduling, that is, the allocation of distributed computational resources to user applications, is one of the most challenging and complex task in Grid computing. The problem of allocating resources in Grid scheduling requires the definition of a model that allows local and external schedulers to communicate in order to achieve an efficient management of the resources themselves. To this aim, some economic/market-based models have been introduced in the literature, where users, external schedulers, and local schedulers negotiate to optimize their objectives. In this paper, we propose a tender/contract-net model for Grid resource allocation, showing the interactions among the involved actors. The performance of the proposed market-based approach is experimentally compared with a round-robin allocation protocol.
... 3 Scenarios for the Use of Grids in Satellite Image Processing In [17] [18] ...
... Scenario 2: Remote Parallel Processing of Satellite Images [18]. The conditions and requirements are the same as in Scenario 1, but, in addition, neither the local nor the remotely accessible computing nodes can individually do the task in a reasonable time. ...
Remote sensing image processing is a very demanding procedure in terms of data manipulation and computing power. Grid computing is a possible solution when the required computing performance or data sharing is not available at the user's site. Two scenarios of using Service Grids were analyzed in our papers [17, 18]. This paper discusses another scenario of using Computational Grids. According to this scenario a prototype code for satellite image classification was designed, implemented and tested.
