Auto-Scheduling: control flow and data flow come together
Abstract: "This paper presents a framework we term auto- scheduling, which brings together the control flow and data flow models by combining most of the advantages and excluding the major disadvantages of the two familiar models. Auto-scheduling can be viewed either as an abstract architectura...
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Buch |
| Sprache: | English |
| Veröffentlicht: |
Urbana, Ill.
1990
|
| Schriftenreihe: | Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report
1058 |
| Schlagworte: | |
| Zusammenfassung: | Abstract: "This paper presents a framework we term auto- scheduling, which brings together the control flow and data flow models by combining most of the advantages and excluding the major disadvantages of the two familiar models. Auto-scheduling can be viewed either as an abstract architectural model or as a parallel program compilation framework. While in ordinary environments parallel task creation and scheduling is done by the operating system, or at best the run-time library, in auto-scheduling task creation and scheduling is performed by the user program itself, making parallel processing affordable at fine- granularity levels Under auto-scheduling the compiler does not only generate object code, but it 'lends' its knowledge about a program to the parallel instruction threads of that program, allowing them to manage, activate, and schedule themselves at run-time, without the need of an external monitor. This is done by means of special drive-code injected by the compiler to each schedulable unit of a program (task, thread, etc). We argue that auto-scheduling offers an optimal approach for exploiting parallelism on real parallel computer systems. |
| Beschreibung: | 28 S. |
Internformat
MARC
| LEADER | 00000nam a2200000 cb4500 | ||
|---|---|---|---|
| 001 | BV008979071 | ||
| 003 | DE-604 | ||
| 005 | 00000000000000.0 | ||
| 007 | t | ||
| 008 | 940206s1990 |||| 00||| eng d | ||
| 035 | |a (OCoLC)25616924 | ||
| 035 | |a (DE-599)BVBBV008979071 | ||
| 040 | |a DE-604 |b ger |e rakddb | ||
| 041 | 0 | |a eng | |
| 049 | |a DE-29T | ||
| 100 | 1 | |a Polychronopoulos, Constantine D. |d 1958- |e Verfasser |0 (DE-588)120999773 |4 aut | |
| 245 | 1 | 0 | |a Auto-Scheduling |b control flow and data flow come together |c Constantine D. Polychronopoulos |
| 264 | 1 | |a Urbana, Ill. |c 1990 | |
| 300 | |a 28 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 1058 | |
| 520 | 3 | |a Abstract: "This paper presents a framework we term auto- scheduling, which brings together the control flow and data flow models by combining most of the advantages and excluding the major disadvantages of the two familiar models. Auto-scheduling can be viewed either as an abstract architectural model or as a parallel program compilation framework. While in ordinary environments parallel task creation and scheduling is done by the operating system, or at best the run-time library, in auto-scheduling task creation and scheduling is performed by the user program itself, making parallel processing affordable at fine- granularity levels | |
| 520 | 3 | |a Under auto-scheduling the compiler does not only generate object code, but it 'lends' its knowledge about a program to the parallel instruction threads of that program, allowing them to manage, activate, and schedule themselves at run-time, without the need of an external monitor. This is done by means of special drive-code injected by the compiler to each schedulable unit of a program (task, thread, etc). We argue that auto-scheduling offers an optimal approach for exploiting parallelism on real parallel computer systems. | |
| 650 | 4 | |a Parallel processing (Electronic computers) | |
| 650 | 4 | |a Production scheduling | |
| 830 | 0 | |a Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report |v 1058 |w (DE-604)BV008930033 |9 1058 | |
| 999 | |a oai:aleph.bib-bvb.de:BVB01-005929831 | ||
Datensatz im Suchindex
| _version_ | 1804123317847719936 |
|---|---|
| any_adam_object | |
| author | Polychronopoulos, Constantine D. 1958- |
| author_GND | (DE-588)120999773 |
| author_facet | Polychronopoulos, Constantine D. 1958- |
| author_role | aut |
| author_sort | Polychronopoulos, Constantine D. 1958- |
| author_variant | c d p cd cdp |
| building | Verbundindex |
| bvnumber | BV008979071 |
| ctrlnum | (OCoLC)25616924 (DE-599)BVBBV008979071 |
| format | Book |
| fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>02242nam a2200313 cb4500</leader><controlfield tag="001">BV008979071</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">00000000000000.0</controlfield><controlfield tag="007">t</controlfield><controlfield tag="008">940206s1990 |||| 00||| eng d</controlfield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)25616924</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV008979071</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">Polychronopoulos, Constantine D.</subfield><subfield code="d">1958-</subfield><subfield code="e">Verfasser</subfield><subfield code="0">(DE-588)120999773</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Auto-Scheduling</subfield><subfield code="b">control flow and data flow come together</subfield><subfield code="c">Constantine D. Polychronopoulos</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">28 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">1058</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Abstract: "This paper presents a framework we term auto- scheduling, which brings together the control flow and data flow models by combining most of the advantages and excluding the major disadvantages of the two familiar models. Auto-scheduling can be viewed either as an abstract architectural model or as a parallel program compilation framework. While in ordinary environments parallel task creation and scheduling is done by the operating system, or at best the run-time library, in auto-scheduling task creation and scheduling is performed by the user program itself, making parallel processing affordable at fine- granularity levels</subfield></datafield><datafield tag="520" ind1="3" ind2=" "><subfield code="a">Under auto-scheduling the compiler does not only generate object code, but it 'lends' its knowledge about a program to the parallel instruction threads of that program, allowing them to manage, activate, and schedule themselves at run-time, without the need of an external monitor. This is done by means of special drive-code injected by the compiler to each schedulable unit of a program (task, thread, etc). We argue that auto-scheduling offers an optimal approach for exploiting parallelism on real parallel computer systems.</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Parallel processing (Electronic computers)</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Production scheduling</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">1058</subfield><subfield code="w">(DE-604)BV008930033</subfield><subfield code="9">1058</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-005929831</subfield></datafield></record></collection> |
| id | DE-604.BV008979071 |
| illustrated | Not Illustrated |
| indexdate | 2024-07-09T17:27:51Z |
| institution | BVB |
| language | English |
| oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-005929831 |
| oclc_num | 25616924 |
| open_access_boolean | |
| owner | DE-29T |
| owner_facet | DE-29T |
| physical | 28 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 | Polychronopoulos, Constantine D. 1958- Verfasser (DE-588)120999773 aut Auto-Scheduling control flow and data flow come together Constantine D. Polychronopoulos Urbana, Ill. 1990 28 S. txt rdacontent n rdamedia nc rdacarrier Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 1058 Abstract: "This paper presents a framework we term auto- scheduling, which brings together the control flow and data flow models by combining most of the advantages and excluding the major disadvantages of the two familiar models. Auto-scheduling can be viewed either as an abstract architectural model or as a parallel program compilation framework. While in ordinary environments parallel task creation and scheduling is done by the operating system, or at best the run-time library, in auto-scheduling task creation and scheduling is performed by the user program itself, making parallel processing affordable at fine- granularity levels Under auto-scheduling the compiler does not only generate object code, but it 'lends' its knowledge about a program to the parallel instruction threads of that program, allowing them to manage, activate, and schedule themselves at run-time, without the need of an external monitor. This is done by means of special drive-code injected by the compiler to each schedulable unit of a program (task, thread, etc). We argue that auto-scheduling offers an optimal approach for exploiting parallelism on real parallel computer systems. Parallel processing (Electronic computers) Production scheduling Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report 1058 (DE-604)BV008930033 1058 |
| spellingShingle | Polychronopoulos, Constantine D. 1958- Auto-Scheduling control flow and data flow come together Center for Supercomputing Research and Development <Urbana, Ill.>: CSRD report Parallel processing (Electronic computers) Production scheduling |
| title | Auto-Scheduling control flow and data flow come together |
| title_auth | Auto-Scheduling control flow and data flow come together |
| title_exact_search | Auto-Scheduling control flow and data flow come together |
| title_full | Auto-Scheduling control flow and data flow come together Constantine D. Polychronopoulos |
| title_fullStr | Auto-Scheduling control flow and data flow come together Constantine D. Polychronopoulos |
| title_full_unstemmed | Auto-Scheduling control flow and data flow come together Constantine D. Polychronopoulos |
| title_short | Auto-Scheduling |
| title_sort | auto scheduling control flow and data flow come together |
| title_sub | control flow and data flow come together |
| topic | Parallel processing (Electronic computers) Production scheduling |
| topic_facet | Parallel processing (Electronic computers) Production scheduling |
| volume_link | (DE-604)BV008930033 |
| work_keys_str_mv | AT polychronopoulosconstantined autoschedulingcontrolflowanddataflowcometogether |