Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods:
Abstract: "The objective of this work is to map certain algorithms for the efficient solution of a class of elliptic partial differential equations on an architecture with multiple levels of computational parallelism and hierarchical memory organization. The prototype system is the Cedar multic...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Buch |
| Sprache: | English |
| Veröffentlicht: |
Urbana, Ill.
1990
|
| Schriftenreihe: | Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report
1056 |
| Schlagworte: | |
| Zusammenfassung: | Abstract: "The objective of this work is to map certain algorithms for the efficient solution of a class of elliptic partial differential equations on an architecture with multiple levels of computational parallelism and hierarchical memory organization. The prototype system is the Cedar multicluster, developed at the Center for Supercomputing Research and Development (CSRD) at the University of Illinois. In our experiments, each cluster of Cedar is a modified Alliant FX/8 with four processors. The algorithms we implement are parallel block cyclic reduction, and certain domain decomposition methods based on the Schur complement, for T-shaped regions They are written to exploit the multiple cluster capabilities of the machine, as well as the concurrent and vector computational resources of each cluster. For each of these applications we obtain good multi- cluster performance: good speedups are obtained for the new implementations on multiple clusters of Cedar, relative to the implementations best suited for a single cluster. Most importantly, our experiments also point to advantages of the multi-cluster organization. For the case of block cyclic reduction, for example, the performance using the new multicluster algorithm on eight processors organized in two clusters is superior to that of the original parallel algorithm optimized for the eight faster processors of the Alliant FX/80. |
| Beschreibung: | Urbana, Ill., Univ., Diss. |
| Beschreibung: | 69 S. |
Internformat
MARC
| LEADER | 00000nam a2200000 cb4500 | ||
|---|---|---|---|
| 001 | BV008979072 | ||
| 003 | DE-604 | ||
| 005 | 00000000000000.0 | ||
| 007 | t | ||
| 008 | 940206s1990 m||| 00||| eng d | ||
| 035 | |a (OCoLC)25298245 | ||
| 035 | |a (DE-599)BVBBV008979072 | ||
| 040 | |a DE-604 |b ger |e rakddb | ||
| 041 | 0 | |a eng | |
| 049 | |a DE-29T | ||
| 100 | 1 | |a Frank, George N. |e Verfasser |4 aut | |
| 245 | 1 | 0 | |a Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |c by George N. Frank |
| 246 | 1 | 3 | |a Reportnr.: UILU ENG 90 8032 |
| 264 | 1 | |a Urbana, Ill. |c 1990 | |
| 300 | |a 69 S. | ||
| 336 | |b txt |2 rdacontent | ||
| 337 | |b n |2 rdamedia | ||
| 338 | |b nc |2 rdacarrier | ||
| 490 | 1 | |a Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report |v 1056 | |
| 500 | |a Urbana, Ill., Univ., Diss. | ||
| 520 | 3 | |a Abstract: "The objective of this work is to map certain algorithms for the efficient solution of a class of elliptic partial differential equations on an architecture with multiple levels of computational parallelism and hierarchical memory organization. The prototype system is the Cedar multicluster, developed at the Center for Supercomputing Research and Development (CSRD) at the University of Illinois. In our experiments, each cluster of Cedar is a modified Alliant FX/8 with four processors. The algorithms we implement are parallel block cyclic reduction, and certain domain decomposition methods based on the Schur complement, for T-shaped regions | |
| 520 | 3 | |a They are written to exploit the multiple cluster capabilities of the machine, as well as the concurrent and vector computational resources of each cluster. For each of these applications we obtain good multi- cluster performance: good speedups are obtained for the new implementations on multiple clusters of Cedar, relative to the implementations best suited for a single cluster. Most importantly, our experiments also point to advantages of the multi-cluster organization. For the case of block cyclic reduction, for example, the performance using the new multicluster algorithm on eight processors organized in two clusters is superior to that of the original parallel algorithm optimized for the eight faster processors of the Alliant FX/80. | |
| 650 | 4 | |a Algorithms | |
| 650 | 4 | |a Differential equations, Partial | |
| 650 | 4 | |a Parallel processing (Electronic computers) | |
| 655 | 7 | |0 (DE-588)4113937-9 |a Hochschulschrift |2 gnd-content | |
| 830 | 0 | |a Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report |v 1056 |w (DE-604)BV008930033 |9 1056 | |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-005929832 | ||
Datensatz im Suchindex
| _version_ | 1804123317851914240 |
|---|---|
| any_adam_object | |
| author | Frank, George N. |
| author_facet | Frank, George N. |
| author_role | aut |
| author_sort | Frank, George N. |
| author_variant | g n f gn gnf |
| building | Verbundindex |
| bvnumber | BV008979072 |
| ctrlnum | (OCoLC)25298245 (DE-599)BVBBV008979072 |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>02683nam a2200361 cb4500</leader><controlfield tag="001">BV008979072</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">00000000000000.0</controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">940206s1990 m||| 00||| eng d</controlfield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)25298245</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV008979072</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rakddb</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="049" ind1=" " ind2=" "><subfield code="a">DE-29T</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Frank, George N.</subfield><subfield code="e">Verfasser</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods</subfield><subfield code="c">by George N. Frank</subfield></datafield><datafield tag="246" ind1="1" ind2="3"><subfield code="a">Reportnr.: UILU ENG 90 8032</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Urbana, Ill.</subfield><subfield code="c">1990</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">69 S.</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="490" ind1="1" ind2=" "><subfield code="a">Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report</subfield><subfield code="v">1056</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">Urbana, Ill., Univ., Diss.</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Abstract: "The objective of this work is to map certain algorithms for the efficient solution of a class of elliptic partial differential equations on an architecture with multiple levels of computational parallelism and hierarchical memory organization. The prototype system is the Cedar multicluster, developed at the Center for Supercomputing Research and Development (CSRD) at the University of Illinois. In our experiments, each cluster of Cedar is a modified Alliant FX/8 with four processors. The algorithms we implement are parallel block cyclic reduction, and certain domain decomposition methods based on the Schur complement, for T-shaped regions</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">They are written to exploit the multiple cluster capabilities of the machine, as well as the concurrent and vector computational resources of each cluster. For each of these applications we obtain good multi- cluster performance: good speedups are obtained for the new implementations on multiple clusters of Cedar, relative to the implementations best suited for a single cluster. Most importantly, our experiments also point to advantages of the multi-cluster organization. For the case of block cyclic reduction, for example, the performance using the new multicluster algorithm on eight processors organized in two clusters is superior to that of the original parallel algorithm optimized for the eight faster processors of the Alliant FX/80.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Algorithms</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Differential equations, Partial</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Parallel processing (Electronic computers)</subfield></datafield><datafield tag="655" ind1=" " ind2="7"><subfield code="0">(DE-588)4113937-9</subfield><subfield code="a">Hochschulschrift</subfield><subfield code="2">gnd-content</subfield></datafield><datafield tag="830" ind1=" " ind2="0"><subfield code="a">Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report</subfield><subfield code="v">1056</subfield><subfield code="w">(DE-604)BV008930033</subfield><subfield code="9">1056</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-005929832</subfield></datafield></record></collection> |
| genre | (DE-588)4113937-9 Hochschulschrift gnd-content |
| genre_facet | Hochschulschrift |
| id | DE-604.BV008979072 |
| illustrated | Not Illustrated |
| indexdate | 2024-07-09T17:27:51Z |
| institution | BVB |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-005929832 |
| oclc_num | 25298245 |
| open_access_boolean | |
| owner | DE-29T |
| owner_facet | DE-29T |
| physical | 69 S. |
| publishDate | 1990 |
| publishDateSearch | 1990 |
| publishDateSort | 1990 |
| record_format | marc |
| series | Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report |
| series2 | Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report |
| spelling | Frank, George N. Verfasser aut Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods by George N. Frank Reportnr.: UILU ENG 90 8032 Urbana, Ill. 1990 69 S. txt rdacontent n rdamedia nc rdacarrier Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 1056 Urbana, Ill., Univ., Diss. Abstract: "The objective of this work is to map certain algorithms for the efficient solution of a class of elliptic partial differential equations on an architecture with multiple levels of computational parallelism and hierarchical memory organization. The prototype system is the Cedar multicluster, developed at the Center for Supercomputing Research and Development (CSRD) at the University of Illinois. In our experiments, each cluster of Cedar is a modified Alliant FX/8 with four processors. The algorithms we implement are parallel block cyclic reduction, and certain domain decomposition methods based on the Schur complement, for T-shaped regions They are written to exploit the multiple cluster capabilities of the machine, as well as the concurrent and vector computational resources of each cluster. For each of these applications we obtain good multi- cluster performance: good speedups are obtained for the new implementations on multiple clusters of Cedar, relative to the implementations best suited for a single cluster. Most importantly, our experiments also point to advantages of the multi-cluster organization. For the case of block cyclic reduction, for example, the performance using the new multicluster algorithm on eight processors organized in two clusters is superior to that of the original parallel algorithm optimized for the eight faster processors of the Alliant FX/80. Algorithms Differential equations, Partial Parallel processing (Electronic computers) (DE-588)4113937-9 Hochschulschrift gnd-content Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 1056 (DE-604)BV008930033 1056 |
| spellingShingle | Frank, George N. Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report Algorithms Differential equations, Partial Parallel processing (Electronic computers) |
| subject_GND | (DE-588)4113937-9 |
| title | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |
| title_alt | Reportnr.: UILU ENG 90 8032 |
| title_auth | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |
| title_exact_search | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |
| title_full | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods by George N. Frank |
| title_fullStr | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods by George N. Frank |
| title_full_unstemmed | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods by George N. Frank |
| title_short | Experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |
| title_sort | experiments on the cedar multicluster with parallel block cyclic reduction and an application to domain decomposition methods |
| topic | Algorithms Differential equations, Partial Parallel processing (Electronic computers) |
| topic_facet | Algorithms Differential equations, Partial Parallel processing (Electronic computers) Hochschulschrift |
| volume_link | (DE-604)BV008930033 |
| work_keys_str_mv | AT frankgeorgen experimentsonthecedarmulticlusterwithparallelblockcyclicreductionandanapplicationtodomaindecompositionmethods AT frankgeorgen reportnruilueng908032 |