Composing Low-Overhead Scheduling Strategies for Improving Performance of Scientific Applications
Many different sources of overheads impact the efficiency of a scheduling strategy applied to a parallel loop within a scientific application. In prior work, we handled these overheads using multiple loop scheduling strategies, with each scheduling strategy focusing on mitigating a subset of the ove...
Saved in:
Published in | OpenMP: Heterogenous Execution and Data Movements pp. 18 - 29 |
---|---|
Main Authors | , |
Format | Book Chapter |
Language | English |
Published |
Cham
Springer International Publishing
|
Series | Lecture Notes in Computer Science |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Many different sources of overheads impact the efficiency of a scheduling strategy applied to a parallel loop within a scientific application. In prior work, we handled these overheads using multiple loop scheduling strategies, with each scheduling strategy focusing on mitigating a subset of the overheads. However, mitigating the impact of one source of overhead can lead to an increase in the impact of another source of overhead, and vice versa. In this work, we show that in order to improve efficiency of loop scheduling strategies, one must adapt the loop scheduling strategies so as to handle all overheads simultaneously. To show this, we describe a composition of our existing loop scheduling strategies, and experiment with the composed scheduling strategy on standard benchmarks and application codes. Applying the composed scheduling strategy to three MPI+OpenMP scientific codes run on a cluster of SMPs improves performance an average of 31 % over standard OpenMP static scheduling. |
---|---|
AbstractList | Many different sources of overheads impact the efficiency of a scheduling strategy applied to a parallel loop within a scientific application. In prior work, we handled these overheads using multiple loop scheduling strategies, with each scheduling strategy focusing on mitigating a subset of the overheads. However, mitigating the impact of one source of overhead can lead to an increase in the impact of another source of overhead, and vice versa. In this work, we show that in order to improve efficiency of loop scheduling strategies, one must adapt the loop scheduling strategies so as to handle all overheads simultaneously. To show this, we describe a composition of our existing loop scheduling strategies, and experiment with the composed scheduling strategy on standard benchmarks and application codes. Applying the composed scheduling strategy to three MPI+OpenMP scientific codes run on a cluster of SMPs improves performance an average of 31 % over standard OpenMP static scheduling. |
Author | Gropp, William D. Kale, Vivek |
Author_xml | – sequence: 1 givenname: Vivek surname: Kale fullname: Kale, Vivek email: vivek@illinois.edu organization: University of Illinois at Urbana-Champaign, Urbana, USA – sequence: 2 givenname: William D. surname: Gropp fullname: Gropp, William D. organization: University of Illinois at Urbana-Champaign, Urbana, USA |
BookMark | eNo1kN1OwzAMhQMMiW3sCbjpCwSSOmmSy2niZ9KkIQ2uozR1t8LWVE0Zr082wDdHPse2rG9CRm1okZA7zu45Y-rBKE2BAjc0F9JIamx-QSaQjHMPl2TMC84pgDBXZJbG_zORj8iYAcupUQJuyCzGD8YYl1qCNmPiFuHQhdi022wVvun6iP0OXZVt_A6rr_3J3wy9G3DbYMzq0GfLQ9eH4yl4xT4ZB9d6zEKdVhpsh6ZufDbvun3j3dCENt6S69rtI87-dErenx7fFi90tX5eLuYrGrlRA-W6dFhorXjldV77SimnpBBeCid8UahULpfIlYGyAOZcpRFLKFHyGtDDlPDfu7Hr03fY2zKEz2g5syeGNkGxYBMWe2Zmk8IP_8Njkw |
ContentType | Book Chapter |
Copyright | Springer International Publishing Switzerland 2015 |
Copyright_xml | – notice: Springer International Publishing Switzerland 2015 |
DOI | 10.1007/978-3-319-24595-9_2 |
DatabaseTitleList | |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering Computer Science |
EISBN | 3319245953 9783319245959 |
EISSN | 1611-3349 |
Editor | Terboven, Christian Müller, Matthias S. Chapman, Barbara M. de Supinski, Bronis R. Reble, Pablo |
Editor_xml | – sequence: 1 givenname: Christian surname: Terboven fullname: Terboven, Christian email: terboven@itc.rwth-aachen.de – sequence: 2 givenname: Bronis R. surname: de Supinski fullname: de Supinski, Bronis R. email: bronis@llnl.gov – sequence: 3 givenname: Pablo surname: Reble fullname: Reble, Pablo email: reble@itc.rwth-aachen.de – sequence: 4 givenname: Barbara M. surname: Chapman fullname: Chapman, Barbara M. email: chapman@cs.uh.edu – sequence: 5 givenname: Matthias S. surname: Müller fullname: Müller, Matthias S. email: mueller@rz.rwth-aachen.de |
EndPage | 29 |
GroupedDBID | -DT -GH -~X 1SB 29L 2HA 2HV 5QI 875 AASHB ABMNI ACGFS ADCXD AEFIE ALMA_UNASSIGNED_HOLDINGS EJD F5P FEDTE HVGLF LAS LDH P2P RIG RNI RSU SVGTG VI1 ~02 |
ID | FETCH-LOGICAL-s197t-18bae68871dc82fcd77a7544c54a4c667777a25e1793b630aad8eeb3be51f3ec3 |
ISBN | 9783319245942 3319245945 |
ISSN | 0302-9743 |
IngestDate | Tue Oct 01 19:45:24 EDT 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
LinkModel | OpenURL |
MergedId | FETCHMERGED-LOGICAL-s197t-18bae68871dc82fcd77a7544c54a4c667777a25e1793b630aad8eeb3be51f3ec3 |
PageCount | 12 |
ParticipantIDs | springer_books_10_1007_978_3_319_24595_9_2 |
PublicationPlace | Cham |
PublicationPlace_xml | – name: Cham |
PublicationSeriesSubtitle | Programming and Software Engineering |
PublicationSeriesTitle | Lecture Notes in Computer Science |
PublicationSeriesTitleAlternate | Lect.Notes Computer |
PublicationSubtitle | 11th International Workshop on OpenMP, IWOMP 2015, Aachen, Germany, October 1-2, 2015, Proceedings |
PublicationTitle | OpenMP: Heterogenous Execution and Data Movements |
Publisher | Springer International Publishing |
Publisher_xml | – name: Springer International Publishing |
RelatedPersons | Kleinberg, Jon M. Mattern, Friedemann Naor, Moni Mitchell, John C. Terzopoulos, Demetri Steffen, Bernhard Pandu Rangan, C. Kanade, Takeo Kittler, Josef Weikum, Gerhard Hutchison, David Tygar, Doug |
RelatedPersons_xml | – sequence: 1 givenname: David surname: Hutchison fullname: Hutchison, David organization: Lancaster University, Lancaster, United Kingdom – sequence: 2 givenname: Takeo surname: Kanade fullname: Kanade, Takeo organization: Carnegie Mellon University, Pittsburgh, USA – sequence: 3 givenname: Josef surname: Kittler fullname: Kittler, Josef organization: University of Surrey, Guildford, United Kingdom – sequence: 4 givenname: Jon M. surname: Kleinberg fullname: Kleinberg, Jon M. organization: Cornell University, Ithaca, USA – sequence: 5 givenname: Friedemann surname: Mattern fullname: Mattern, Friedemann organization: CNB H 104.2, ETH Zurich, Zürich, Switzerland – sequence: 6 givenname: John C. surname: Mitchell fullname: Mitchell, John C. organization: Stanford, USA – sequence: 7 givenname: Moni surname: Naor fullname: Naor, Moni organization: Weizmann Institute of Science, Rehovot, Israel – sequence: 8 givenname: C. surname: Pandu Rangan fullname: Pandu Rangan, C. organization: Madras, Indian Institute of Technology, Chennai, India – sequence: 9 givenname: Bernhard surname: Steffen fullname: Steffen, Bernhard organization: Fakultät Informatik, TU Dortmund, Dortmund, Germany – sequence: 10 givenname: Demetri surname: Terzopoulos fullname: Terzopoulos, Demetri organization: University of California, Los Angeles, USA – sequence: 11 givenname: Doug surname: Tygar fullname: Tygar, Doug organization: University of California, Berkeley, USA – sequence: 12 givenname: Gerhard surname: Weikum fullname: Weikum, Gerhard organization: Max Planck Institute for Informatics, Saarbrücken, Germany |
SSID | ssj0001585389 ssj0002792 |
Score | 1.9493386 |
Snippet | Many different sources of overheads impact the efficiency of a scheduling strategy applied to a parallel loop within a scientific application. In prior work,... |
SourceID | springer |
SourceType | Publisher |
StartPage | 18 |
SubjectTerms | Dynamic Fraction Loop Iteration Performance Loss Schedule Strategy Static Fraction |
Title | Composing Low-Overhead Scheduling Strategies for Improving Performance of Scientific Applications |
URI | http://link.springer.com/10.1007/978-3-319-24595-9_2 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1Bb9MwFLbKOMEBGCA2AfKBE5MnUsd2euCAoFBN3dhhm3aLHOfl2ExrKhi_nvdsJ_FaLqOHtIqiJH1f8t7z5_c-M_ahbooCauNErq0Rea1BWAlaYO6gcg2NM15d__RMLy7zk2t1PZncJVVLm646dn_-2VfyP6jiPsSVumQfgOxwUtyBvxFf3CLCuN1Kfu_TrGGttRtYnZ7TiH5BJS1tVFud_wa36YYiY9tZfG-9KHg3ztvEIuIrdHVDp86P2_bGL1YXKZijb8fp80SOo_XEwrL9JX6irdCN16TiicEq6nr3shO-dnGkK86T5gTqkfE9mFSiFHLglDQks8H68zLObJy1nS8YO-oXn-h90Q4xuUVtjuzavZGslDQSVLM8JTslemsc7wQHCMFBa5JdlEHmNDrd6MBD-A78yU5gSGtBqG-LrqXErMTo_cjM0B0-_jI_WV6N9BwOo_y68DGok85imJAKt0RtQv0tqyDkNP6FQd0qCBhvXXFnzt2nMhfP2VNqb-HUd4IGe8EmsNpnz3r78mjfffYkUap8yewAP0_h5yP8fISfI9p8gJ8n8PO24SP8PIX_Fbv8Pr_4uhBxdQ6xzmamE1lRWdAYo7LaFdPG1cZYUlN0Kre509rgx04VUASotPxkbV0AVLIClTUSnHzN9lbtCt4w3kgFUwNFZjKSK4TCVbZQdiodRghZNQfsY2-xkt63ddmLbaN5S1mieUtv3hK_Dx9y8Fu2191u4B0mll31Pj4EfwGs63g7 |
link.rule.ids | 785,786,790,799,27958 |
linkProvider | Library Specific Holdings |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.title=OpenMP%3A+Heterogenous+Execution+and+Data+Movements&rft.au=Kale%2C+Vivek&rft.au=Gropp%2C+William+D.&rft.atitle=Composing+Low-Overhead+Scheduling+Strategies+for+Improving+Performance+of+Scientific+Applications&rft.series=Lecture+Notes+in+Computer+Science&rft.pub=Springer+International+Publishing&rft.isbn=9783319245942&rft.issn=0302-9743&rft.eissn=1611-3349&rft.spage=18&rft.epage=29&rft_id=info:doi/10.1007%2F978-3-319-24595-9_2 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0302-9743&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0302-9743&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0302-9743&client=summon |