Show simple item record

dc.identifier.urihttp://hdl.handle.net/1951/55586
dc.identifier.urihttp://hdl.handle.net/11401/72637
dc.description.sponsorshipThis work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.en_US
dc.formatMonograph
dc.format.mediumElectronic Resourceen_US
dc.language.isoen_US
dc.publisherThe Graduate School, Stony Brook University: Stony Brook, NY.
dc.typeDissertation
dcterms.abstractAs supercomputers have reached nearly 300,000 cores with 2 petaflops in Linpack performance in June 2010, energy consumption and temperature control are posing a developmental bottleneck. We retrospectively and comparatively examine all of the available data contained in the Green500 list that launched in November 2007, and the Top500 list, and propose a novel representation and analysis of the data, highlighting major evolutionary trends.With these new insights, we introduce a new technique for generating more efficient networks by systematically interlacing bypass rings to torus networks (iBT networks). The resulting network can improve the original torus network by reducing the network diameter, node-to-node distances, and by increasing the bisection width without increasing wiring and other engineering complexity. We present and analyze the statement that a 3D iBT network proposed by our technique outperforms 4D torus networks of the same node degree. We found that interlacing rings of sizes 6 and 12 to all three dimensions of a torus network with meshes 30 30 36 generates the best network of all possible networks, including 4D torus and hypercube of approximately 32,000 nodes. This demonstrates that strategically interlacing bypass rings into a 3D torus network enhances the torus network more effectively than adding a fourth dimension, although we may generalize the claim. We also present a node-to-node distance formula for the iBT networks.
dcterms.available2012-05-15T18:05:58Z
dcterms.available2015-04-24T14:52:57Z
dcterms.contributorDeng, Yuefanen_US
dcterms.contributorBrent Lindquisten_US
dcterms.contributorAlan Tuckeren_US
dcterms.contributorDantong Yu.en_US
dcterms.creatorPowell, Reid
dcterms.dateAccepted2012-05-15T18:05:58Z
dcterms.dateAccepted2015-04-24T14:52:57Z
dcterms.dateSubmitted2012-05-15T18:05:58Z
dcterms.dateSubmitted2015-04-24T14:52:57Z
dcterms.descriptionDepartment of Applied Mathematics and Statisticsen_US
dcterms.formatApplication/PDFen_US
dcterms.formatMonograph
dcterms.identifierhttp://hdl.handle.net/1951/55586
dcterms.identifierPowell_grad.sunysb_0771E_10407.pdfen_US
dcterms.identifierhttp://hdl.handle.net/11401/72637
dcterms.issued2010-12-01
dcterms.languageen_US
dcterms.provenanceMade available in DSpace on 2012-05-15T18:05:58Z (GMT). No. of bitstreams: 1 Powell_grad.sunysb_0771E_10407.pdf: 4516614 bytes, checksum: 3f5e33c0d6c79bfb5831bfe22b5421be (MD5) Previous issue date: 1en
dcterms.provenanceMade available in DSpace on 2015-04-24T14:52:57Z (GMT). No. of bitstreams: 3 Powell_grad.sunysb_0771E_10407.pdf.jpg: 1894 bytes, checksum: a6009c46e6ec8251b348085684cba80d (MD5) Powell_grad.sunysb_0771E_10407.pdf.txt: 104389 bytes, checksum: 27db748cf944d3c6f85e9939a40d865f (MD5) Powell_grad.sunysb_0771E_10407.pdf: 4516614 bytes, checksum: 3f5e33c0d6c79bfb5831bfe22b5421be (MD5) Previous issue date: 1en
dcterms.publisherThe Graduate School, Stony Brook University: Stony Brook, NY.
dcterms.subjectApplied Mathematics
dcterms.titleAnalysis of Supercomputers and Development of a Novel Network
dcterms.typeDissertation


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record