Practical Size-based Scheduling for MapReduce Workloads
We present the Hadoop Fair Sojourn Protocol (HFSP) scheduler, which implements a size-based scheduling discipline for Hadoop. The benefits of size-based scheduling disciplines are well recognized in a variety of contexts (computer networks, operating systems, etc...), yet, their practical implementa...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | |
container_volume | |
creator | Pastorelli, Mario Barbuzzi, Antonio Carra, Damiano Dell'Amico, Matteo Michiardi, Pietro |
description | We present the Hadoop Fair Sojourn Protocol (HFSP) scheduler, which
implements a size-based scheduling discipline for Hadoop. The benefits of
size-based scheduling disciplines are well recognized in a variety of contexts
(computer networks, operating systems, etc...), yet, their practical
implementation for a system such as Hadoop raises a number of important
challenges. With HFSP, which is available as an open-source project, we address
issues related to job size estimation, resource management and study the
effects of a variety of preemption strategies. Although the architecture
underlying HFSP is suitable for any size-based scheduling discipline, in this
work we revisit and extend the Fair Sojourn Protocol, which solves problems
related to job starvation that affect FIFO, Processor Sharing and a range of
size-based disciplines. Our experiments, in which we compare HFSP to standard
Hadoop schedulers, pinpoint at a significant decrease in average job sojourn
times - a metric that accounts for the total time a job spends in the system,
including waiting and serving times - for realistic workloads that we generate
according to production traces available in literature. |
doi_str_mv | 10.48550/arxiv.1302.2749 |
format | Article |
fullrecord | <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1302_2749</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1302_2749</sourcerecordid><originalsourceid>FETCH-LOGICAL-a659-4ecdb5c1065024034f8125d8a0c676af0cfca1835f064c06bfeb826981013c873</originalsourceid><addsrcrecordid>eNotj0tPAjEURrthQdA9K9M_MOPtczpLQ3yQYDRC4nJy57bVhoEhHSXAr0fU1cm3OfkOY1MBpXbGwC3mQ9qXQoEsZaXrMateM9JXIuz4Mp1C0eIQPF_SZ_DfXdp-8Nhn_oy7t59Ngb_3ed316IcrNorYDeH6nxO2erhfzZ6KxcvjfHa3KNCautCBfGtIgDUgNSgdnZDGOwSylcUIFAmFUyaC1QS2jaF10tZOgFDkKjVhN3_a3-PNLqcN5mNzCWguAeoMmwA_WQ</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Practical Size-based Scheduling for MapReduce Workloads</title><source>arXiv.org</source><creator>Pastorelli, Mario ; Barbuzzi, Antonio ; Carra, Damiano ; Dell'Amico, Matteo ; Michiardi, Pietro</creator><creatorcontrib>Pastorelli, Mario ; Barbuzzi, Antonio ; Carra, Damiano ; Dell'Amico, Matteo ; Michiardi, Pietro</creatorcontrib><description>We present the Hadoop Fair Sojourn Protocol (HFSP) scheduler, which
implements a size-based scheduling discipline for Hadoop. The benefits of
size-based scheduling disciplines are well recognized in a variety of contexts
(computer networks, operating systems, etc...), yet, their practical
implementation for a system such as Hadoop raises a number of important
challenges. With HFSP, which is available as an open-source project, we address
issues related to job size estimation, resource management and study the
effects of a variety of preemption strategies. Although the architecture
underlying HFSP is suitable for any size-based scheduling discipline, in this
work we revisit and extend the Fair Sojourn Protocol, which solves problems
related to job starvation that affect FIFO, Processor Sharing and a range of
size-based disciplines. Our experiments, in which we compare HFSP to standard
Hadoop schedulers, pinpoint at a significant decrease in average job sojourn
times - a metric that accounts for the total time a job spends in the system,
including waiting and serving times - for realistic workloads that we generate
according to production traces available in literature.</description><identifier>DOI: 10.48550/arxiv.1302.2749</identifier><language>eng</language><subject>Computer Science - Distributed, Parallel, and Cluster Computing</subject><creationdate>2013-02</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,780,885</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1302.2749$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1302.2749$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Pastorelli, Mario</creatorcontrib><creatorcontrib>Barbuzzi, Antonio</creatorcontrib><creatorcontrib>Carra, Damiano</creatorcontrib><creatorcontrib>Dell'Amico, Matteo</creatorcontrib><creatorcontrib>Michiardi, Pietro</creatorcontrib><title>Practical Size-based Scheduling for MapReduce Workloads</title><description>We present the Hadoop Fair Sojourn Protocol (HFSP) scheduler, which
implements a size-based scheduling discipline for Hadoop. The benefits of
size-based scheduling disciplines are well recognized in a variety of contexts
(computer networks, operating systems, etc...), yet, their practical
implementation for a system such as Hadoop raises a number of important
challenges. With HFSP, which is available as an open-source project, we address
issues related to job size estimation, resource management and study the
effects of a variety of preemption strategies. Although the architecture
underlying HFSP is suitable for any size-based scheduling discipline, in this
work we revisit and extend the Fair Sojourn Protocol, which solves problems
related to job starvation that affect FIFO, Processor Sharing and a range of
size-based disciplines. Our experiments, in which we compare HFSP to standard
Hadoop schedulers, pinpoint at a significant decrease in average job sojourn
times - a metric that accounts for the total time a job spends in the system,
including waiting and serving times - for realistic workloads that we generate
according to production traces available in literature.</description><subject>Computer Science - Distributed, Parallel, and Cluster Computing</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2013</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotj0tPAjEURrthQdA9K9M_MOPtczpLQ3yQYDRC4nJy57bVhoEhHSXAr0fU1cm3OfkOY1MBpXbGwC3mQ9qXQoEsZaXrMateM9JXIuz4Mp1C0eIQPF_SZ_DfXdp-8Nhn_oy7t59Ngb_3ed316IcrNorYDeH6nxO2erhfzZ6KxcvjfHa3KNCautCBfGtIgDUgNSgdnZDGOwSylcUIFAmFUyaC1QS2jaF10tZOgFDkKjVhN3_a3-PNLqcN5mNzCWguAeoMmwA_WQ</recordid><startdate>20130212</startdate><enddate>20130212</enddate><creator>Pastorelli, Mario</creator><creator>Barbuzzi, Antonio</creator><creator>Carra, Damiano</creator><creator>Dell'Amico, Matteo</creator><creator>Michiardi, Pietro</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20130212</creationdate><title>Practical Size-based Scheduling for MapReduce Workloads</title><author>Pastorelli, Mario ; Barbuzzi, Antonio ; Carra, Damiano ; Dell'Amico, Matteo ; Michiardi, Pietro</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a659-4ecdb5c1065024034f8125d8a0c676af0cfca1835f064c06bfeb826981013c873</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Computer Science - Distributed, Parallel, and Cluster Computing</topic><toplevel>online_resources</toplevel><creatorcontrib>Pastorelli, Mario</creatorcontrib><creatorcontrib>Barbuzzi, Antonio</creatorcontrib><creatorcontrib>Carra, Damiano</creatorcontrib><creatorcontrib>Dell'Amico, Matteo</creatorcontrib><creatorcontrib>Michiardi, Pietro</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Pastorelli, Mario</au><au>Barbuzzi, Antonio</au><au>Carra, Damiano</au><au>Dell'Amico, Matteo</au><au>Michiardi, Pietro</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Practical Size-based Scheduling for MapReduce Workloads</atitle><date>2013-02-12</date><risdate>2013</risdate><abstract>We present the Hadoop Fair Sojourn Protocol (HFSP) scheduler, which
implements a size-based scheduling discipline for Hadoop. The benefits of
size-based scheduling disciplines are well recognized in a variety of contexts
(computer networks, operating systems, etc...), yet, their practical
implementation for a system such as Hadoop raises a number of important
challenges. With HFSP, which is available as an open-source project, we address
issues related to job size estimation, resource management and study the
effects of a variety of preemption strategies. Although the architecture
underlying HFSP is suitable for any size-based scheduling discipline, in this
work we revisit and extend the Fair Sojourn Protocol, which solves problems
related to job starvation that affect FIFO, Processor Sharing and a range of
size-based disciplines. Our experiments, in which we compare HFSP to standard
Hadoop schedulers, pinpoint at a significant decrease in average job sojourn
times - a metric that accounts for the total time a job spends in the system,
including waiting and serving times - for realistic workloads that we generate
according to production traces available in literature.</abstract><doi>10.48550/arxiv.1302.2749</doi><oa>free_for_read</oa></addata></record> |
fulltext | fulltext_linktorsrc |
identifier | DOI: 10.48550/arxiv.1302.2749 |
ispartof | |
issn | |
language | eng |
recordid | cdi_arxiv_primary_1302_2749 |
source | arXiv.org |
subjects | Computer Science - Distributed, Parallel, and Cluster Computing |
title | Practical Size-based Scheduling for MapReduce Workloads |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-24T14%3A09%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Practical%20Size-based%20Scheduling%20for%20MapReduce%20Workloads&rft.au=Pastorelli,%20Mario&rft.date=2013-02-12&rft_id=info:doi/10.48550/arxiv.1302.2749&rft_dat=%3Carxiv_GOX%3E1302_2749%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true |