Reducing synchronization and scheduling overhead in parallel loops:
Abstract: "This thesis examines the effect of run-time overhead on the performance of parallel loops on shared-memory multiprocessors. Parallel execution time and speedup are examined as a function of both the loop scheduling strategy and the barrier synchronization mechanism. Analytic models f...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Buch |
| Sprache: | English |
| Veröffentlicht: |
Urbana, Ill.
1989
|
| Schriftenreihe: | Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report
922 |
| Schlagworte: | |
| Zusammenfassung: | Abstract: "This thesis examines the effect of run-time overhead on the performance of parallel loops on shared-memory multiprocessors. Parallel execution time and speedup are examined as a function of both the loop scheduling strategy and the barrier synchronization mechanism. Analytic models for parallel loop performance are developed and compared to simulation results, and are found to be in good agreement. The models take into account not only deterministic runtime overhead due to scheduling and synchronization, but also random variations in loop iteration execution times with both normal and exponential distributions These models may be used by an intelligent compiler to determine the optimum scheduling strategy and the optimum number of processors for executing a given parallel loop. Various hardware techniques for supporting efficient synchronization and scheduling operations in multiprocessors are examined. A broadcast network is proposed as a lower-cost alternative to a combining network for supporting efficient synchronization operations. A new special-purpose hardware scheme for supporting ideal barrier synchronization is also presented. This scheme allows very fast barrier synchronization, and is hierarchically expandable to almost arbitrarily large systems. |
| Beschreibung: | Zugl.: Urbana, Ill., Univ., Diss. |
| Beschreibung: | XII, 164 S. |
Internformat
MARC
| LEADER | 00000nam a2200000 cb4500 | ||
|---|---|---|---|
| 001 | BV008949527 | ||
| 003 | DE-604 | ||
| 005 | 00000000000000.0 | ||
| 007 | t | ||
| 008 | 940206s1989 m||| 00||| eng d | ||
| 035 | |a (OCoLC)21941741 | ||
| 035 | |a (DE-599)BVBBV008949527 | ||
| 040 | |a DE-604 |b ger |e rakddb | ||
| 041 | 0 | |a eng | |
| 049 | |a DE-29T | ||
| 100 | 1 | |a Beckmann, Carl J. |e Verfasser |4 aut | |
| 245 | 1 | 0 | |a Reducing synchronization and scheduling overhead in parallel loops |c by Carl Josef Beckmann |
| 246 | 1 | 3 | |a Reportnr. UILU ENG 89 8014 |
| 264 | 1 | |a Urbana, Ill. |c 1989 | |
| 300 | |a XII, 164 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 922 | |
| 500 | |a Zugl.: Urbana, Ill., Univ., Diss. | ||
| 520 | 3 | |a Abstract: "This thesis examines the effect of run-time overhead on the performance of parallel loops on shared-memory multiprocessors. Parallel execution time and speedup are examined as a function of both the loop scheduling strategy and the barrier synchronization mechanism. Analytic models for parallel loop performance are developed and compared to simulation results, and are found to be in good agreement. The models take into account not only deterministic runtime overhead due to scheduling and synchronization, but also random variations in loop iteration execution times with both normal and exponential distributions | |
| 520 | 3 | |a These models may be used by an intelligent compiler to determine the optimum scheduling strategy and the optimum number of processors for executing a given parallel loop. Various hardware techniques for supporting efficient synchronization and scheduling operations in multiprocessors are examined. A broadcast network is proposed as a lower-cost alternative to a combining network for supporting efficient synchronization operations. A new special-purpose hardware scheme for supporting ideal barrier synchronization is also presented. This scheme allows very fast barrier synchronization, and is hierarchically expandable to almost arbitrarily large systems. | |
| 650 | 4 | |a Multiprocessors | |
| 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 922 |w (DE-604)BV008930033 |9 922 | |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-005905184 | ||
Datensatz im Suchindex
| _version_ | 1804123282673238016 |
|---|---|
| any_adam_object | |
| author | Beckmann, Carl J. |
| author_facet | Beckmann, Carl J. |
| author_role | aut |
| author_sort | Beckmann, Carl J. |
| author_variant | c j b cj cjb |
| building | Verbundindex |
| bvnumber | BV008949527 |
| ctrlnum | (OCoLC)21941741 (DE-599)BVBBV008949527 |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>02480nam a2200349 cb4500</leader><controlfield tag="001">BV008949527</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">00000000000000.0</controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">940206s1989 m||| 00||| eng d</controlfield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)21941741</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV008949527</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">Beckmann, Carl J.</subfield><subfield code="e">Verfasser</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Reducing synchronization and scheduling overhead in parallel loops</subfield><subfield code="c">by Carl Josef Beckmann</subfield></datafield><datafield tag="246" ind1="1" ind2="3"><subfield code="a">Reportnr. UILU ENG 89 8014</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Urbana, Ill.</subfield><subfield code="c">1989</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">XII, 164 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">922</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">Zugl.: Urbana, Ill., Univ., Diss.</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Abstract: "This thesis examines the effect of run-time overhead on the performance of parallel loops on shared-memory multiprocessors. Parallel execution time and speedup are examined as a function of both the loop scheduling strategy and the barrier synchronization mechanism. Analytic models for parallel loop performance are developed and compared to simulation results, and are found to be in good agreement. The models take into account not only deterministic runtime overhead due to scheduling and synchronization, but also random variations in loop iteration execution times with both normal and exponential distributions</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">These models may be used by an intelligent compiler to determine the optimum scheduling strategy and the optimum number of processors for executing a given parallel loop. Various hardware techniques for supporting efficient synchronization and scheduling operations in multiprocessors are examined. A broadcast network is proposed as a lower-cost alternative to a combining network for supporting efficient synchronization operations. A new special-purpose hardware scheme for supporting ideal barrier synchronization is also presented. This scheme allows very fast barrier synchronization, and is hierarchically expandable to almost arbitrarily large systems.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Multiprocessors</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">922</subfield><subfield code="w">(DE-604)BV008930033</subfield><subfield code="9">922</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-005905184</subfield></datafield></record></collection> |
| genre | (DE-588)4113937-9 Hochschulschrift gnd-content |
| genre_facet | Hochschulschrift |
| id | DE-604.BV008949527 |
| illustrated | Not Illustrated |
| indexdate | 2024-07-09T17:27:18Z |
| institution | BVB |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-005905184 |
| oclc_num | 21941741 |
| open_access_boolean | |
| owner | DE-29T |
| owner_facet | DE-29T |
| physical | XII, 164 S. |
| publishDate | 1989 |
| publishDateSearch | 1989 |
| publishDateSort | 1989 |
| 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 | Beckmann, Carl J. Verfasser aut Reducing synchronization and scheduling overhead in parallel loops by Carl Josef Beckmann Reportnr. UILU ENG 89 8014 Urbana, Ill. 1989 XII, 164 S. txt rdacontent n rdamedia nc rdacarrier Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 922 Zugl.: Urbana, Ill., Univ., Diss. Abstract: "This thesis examines the effect of run-time overhead on the performance of parallel loops on shared-memory multiprocessors. Parallel execution time and speedup are examined as a function of both the loop scheduling strategy and the barrier synchronization mechanism. Analytic models for parallel loop performance are developed and compared to simulation results, and are found to be in good agreement. The models take into account not only deterministic runtime overhead due to scheduling and synchronization, but also random variations in loop iteration execution times with both normal and exponential distributions These models may be used by an intelligent compiler to determine the optimum scheduling strategy and the optimum number of processors for executing a given parallel loop. Various hardware techniques for supporting efficient synchronization and scheduling operations in multiprocessors are examined. A broadcast network is proposed as a lower-cost alternative to a combining network for supporting efficient synchronization operations. A new special-purpose hardware scheme for supporting ideal barrier synchronization is also presented. This scheme allows very fast barrier synchronization, and is hierarchically expandable to almost arbitrarily large systems. Multiprocessors Parallel processing (Electronic computers) (DE-588)4113937-9 Hochschulschrift gnd-content Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 922 (DE-604)BV008930033 922 |
| spellingShingle | Beckmann, Carl J. Reducing synchronization and scheduling overhead in parallel loops Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report Multiprocessors Parallel processing (Electronic computers) |
| subject_GND | (DE-588)4113937-9 |
| title | Reducing synchronization and scheduling overhead in parallel loops |
| title_alt | Reportnr. UILU ENG 89 8014 |
| title_auth | Reducing synchronization and scheduling overhead in parallel loops |
| title_exact_search | Reducing synchronization and scheduling overhead in parallel loops |
| title_full | Reducing synchronization and scheduling overhead in parallel loops by Carl Josef Beckmann |
| title_fullStr | Reducing synchronization and scheduling overhead in parallel loops by Carl Josef Beckmann |
| title_full_unstemmed | Reducing synchronization and scheduling overhead in parallel loops by Carl Josef Beckmann |
| title_short | Reducing synchronization and scheduling overhead in parallel loops |
| title_sort | reducing synchronization and scheduling overhead in parallel loops |
| topic | Multiprocessors Parallel processing (Electronic computers) |
| topic_facet | Multiprocessors Parallel processing (Electronic computers) Hochschulschrift |
| volume_link | (DE-604)BV008930033 |
| work_keys_str_mv | AT beckmanncarlj reducingsynchronizationandschedulingoverheadinparallelloops AT beckmanncarlj reportnruilueng898014 |