Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/11718
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSingh, Vivek Sheel-
dc.date.accessioned2014-11-28T05:09:24Z-
dc.date.available2014-11-28T05:09:24Z-
dc.date.issued2007-
dc.identifierM.Techen_US
dc.identifier.urihttp://hdl.handle.net/123456789/11718-
dc.guideMisra, Manoj-
dc.description.abstractGrid Computing, which is emerging as a new paradigm for next generation high performance computing, holds a great promise to provide virtual supercomputing environments to solve complex science, engineering and industry application problems, by aggregating geographically remote, heterogeneous resources. Projects Iike SETI@Home, MyGrid, Folding@Home etc. have demonstrated and unleashed the ability and power of computational grids. There is a great demand of computation power in the Petroleum Extraction and Production Industry. The primary use of technical computing cycles in the oil industry is for SDP (seismic data processing) and imaging of the earth's subsurface, driven by the business need for making well-informed drilling decisions by petroleum exploration and production companies. Porting these data processing applications from traditional monolithic supercomputing-machines/clusters to grids would require specialized scheduling algorithms, which take into account the peculiarities of these applications. This report presents the proposed scheduling framework which aims at minimizing the processing time, and therefore maximizing user satisfaction and resource utilization, for SDP applications. The proposed strategy minimizes the network overhead, takes into account job dependencies and performs dynamic job grouping. It, provides User deadline based network ToS; such that critical tasks enjoy expedited Network Service. To ensure maximum Job Completion rate, deadline constrained Time Optimization is refined and integrated with the framework. A Deadline-miss triggered, sender-initiated Task Migration strategy is adopted to balance load efficiently across the resource pool, and to compensate for the inaccuracies at the broker level.en_US
dc.language.isoenen_US
dc.subjectELECTRONICS AND COMPUTER ENGINEERINGen_US
dc.subjectSEISMIC DATA PROCESSINGen_US
dc.subjectGRID SCHEDULING FRAMEWORKen_US
dc.subjectGRID COMPUTINGen_US
dc.titleA GRID SCHEDULING FRAMEWORK FOR SEISMIC DATA PROCESSINGen_US
dc.typeM.Tech Dessertationen_US
dc.accession.numberG13406en_US
Appears in Collections:MASTERS' DISSERTATIONS (E & C)

Files in This Item:
File Description SizeFormat 
ECDG13406.pdf11.12 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.