CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning

Configuration tuning for large software systems is generally challenging due to the complex configuration space and expensive performance evaluation. Most existing approaches follow a two-phase process, first learning a regression-based performance prediction model on available samples and then sear...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cao, Rong, Bao, Liang, Wu, Chase, Zhangsun, Panpan, Li, Yufei, Zhang, Zhe
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 Cao, Rong
Bao, Liang
Wu, Chase
Zhangsun, Panpan
Li, Yufei
Zhang, Zhe
description Configuration tuning for large software systems is generally challenging due to the complex configuration space and expensive performance evaluation. Most existing approaches follow a two-phase process, first learning a regression-based performance prediction model on available samples and then searching for the configurations with satisfactory performance using the learned model. Such regression-based models often suffer from the scarcity of samples due to the enormous time and resources required to run a large software system with a specific configuration. Moreover, previous studies have shown that even a highly accurate regression-based model may fail to discern the relative merit between two configurations, whereas performance comparison is actually one fundamental strategy for configuration tuning. To address these issues, this paper proposes CM-CASL, a Comparison-based performance Modeling approach for software systems via Collaborative Active and Semisupervised Learning. CM-CASL learns a classification model that compares the performance of two given configurations, and enhances the samples through a collaborative labeling process by both human experts and classifiers using an integration of active and semisupervised learning. Experimental results demonstrate that CM-CASL outperforms two state-of-the-art performance modeling approaches in terms of both classification accuracy and rank accuracy, and thus provides a better performance model for the subsequent work of configuration tuning.
doi_str_mv 10.48550/arxiv.2303.15967
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2303_15967</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2303_15967</sourcerecordid><originalsourceid>FETCH-LOGICAL-a677-9f9f24814f3183459bfe2033f960ae7bf7179c6816ed4e8b065e18f32e6e231f3</originalsourceid><addsrcrecordid>eNotkM1OhDAURrtxYUYfwJV9AbCl0BZ3hPiXMNGE2ZML3JomQEmLjPP2zuCszuo7yXcIeeAsTnWWsSfwv3aNE8FEzLNcqltyLPdRWdTVMy3dOIO3wU1RCwF7-oXeOD_C1CHdux4HO31TZ2jtzHIEj7Q-hQXHQFcL5_UwQOs8LHZFWnQbYOppjaMNPzP61V6kFYKfzqI7cmNgCHh_5Y4cXl8O5XtUfb59lEUVgVQqyk1uklTz1AiuRZrlrcGECWFyyQBVaxRXeSc1l9inqFsmM-TaiAQlJoIbsSOP_9rteTN7O4I_NZcCzVZA_AETyVdS</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning</title><source>arXiv.org</source><creator>Cao, Rong ; Bao, Liang ; Wu, Chase ; Zhangsun, Panpan ; Li, Yufei ; Zhang, Zhe</creator><creatorcontrib>Cao, Rong ; Bao, Liang ; Wu, Chase ; Zhangsun, Panpan ; Li, Yufei ; Zhang, Zhe</creatorcontrib><description>Configuration tuning for large software systems is generally challenging due to the complex configuration space and expensive performance evaluation. Most existing approaches follow a two-phase process, first learning a regression-based performance prediction model on available samples and then searching for the configurations with satisfactory performance using the learned model. Such regression-based models often suffer from the scarcity of samples due to the enormous time and resources required to run a large software system with a specific configuration. Moreover, previous studies have shown that even a highly accurate regression-based model may fail to discern the relative merit between two configurations, whereas performance comparison is actually one fundamental strategy for configuration tuning. To address these issues, this paper proposes CM-CASL, a Comparison-based performance Modeling approach for software systems via Collaborative Active and Semisupervised Learning. CM-CASL learns a classification model that compares the performance of two given configurations, and enhances the samples through a collaborative labeling process by both human experts and classifiers using an integration of active and semisupervised learning. Experimental results demonstrate that CM-CASL outperforms two state-of-the-art performance modeling approaches in terms of both classification accuracy and rank accuracy, and thus provides a better performance model for the subsequent work of configuration tuning.</description><identifier>DOI: 10.48550/arxiv.2303.15967</identifier><language>eng</language><subject>Computer Science - Software Engineering</subject><creationdate>2023-03</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/2303.15967$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2303.15967$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Cao, Rong</creatorcontrib><creatorcontrib>Bao, Liang</creatorcontrib><creatorcontrib>Wu, Chase</creatorcontrib><creatorcontrib>Zhangsun, Panpan</creatorcontrib><creatorcontrib>Li, Yufei</creatorcontrib><creatorcontrib>Zhang, Zhe</creatorcontrib><title>CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning</title><description>Configuration tuning for large software systems is generally challenging due to the complex configuration space and expensive performance evaluation. Most existing approaches follow a two-phase process, first learning a regression-based performance prediction model on available samples and then searching for the configurations with satisfactory performance using the learned model. Such regression-based models often suffer from the scarcity of samples due to the enormous time and resources required to run a large software system with a specific configuration. Moreover, previous studies have shown that even a highly accurate regression-based model may fail to discern the relative merit between two configurations, whereas performance comparison is actually one fundamental strategy for configuration tuning. To address these issues, this paper proposes CM-CASL, a Comparison-based performance Modeling approach for software systems via Collaborative Active and Semisupervised Learning. CM-CASL learns a classification model that compares the performance of two given configurations, and enhances the samples through a collaborative labeling process by both human experts and classifiers using an integration of active and semisupervised learning. Experimental results demonstrate that CM-CASL outperforms two state-of-the-art performance modeling approaches in terms of both classification accuracy and rank accuracy, and thus provides a better performance model for the subsequent work of configuration tuning.</description><subject>Computer Science - Software Engineering</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkM1OhDAURrtxYUYfwJV9AbCl0BZ3hPiXMNGE2ZML3JomQEmLjPP2zuCszuo7yXcIeeAsTnWWsSfwv3aNE8FEzLNcqltyLPdRWdTVMy3dOIO3wU1RCwF7-oXeOD_C1CHdux4HO31TZ2jtzHIEj7Q-hQXHQFcL5_UwQOs8LHZFWnQbYOppjaMNPzP61V6kFYKfzqI7cmNgCHh_5Y4cXl8O5XtUfb59lEUVgVQqyk1uklTz1AiuRZrlrcGECWFyyQBVaxRXeSc1l9inqFsmM-TaiAQlJoIbsSOP_9rteTN7O4I_NZcCzVZA_AETyVdS</recordid><startdate>20230328</startdate><enddate>20230328</enddate><creator>Cao, Rong</creator><creator>Bao, Liang</creator><creator>Wu, Chase</creator><creator>Zhangsun, Panpan</creator><creator>Li, Yufei</creator><creator>Zhang, Zhe</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20230328</creationdate><title>CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning</title><author>Cao, Rong ; Bao, Liang ; Wu, Chase ; Zhangsun, Panpan ; Li, Yufei ; Zhang, Zhe</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a677-9f9f24814f3183459bfe2033f960ae7bf7179c6816ed4e8b065e18f32e6e231f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Computer Science - Software Engineering</topic><toplevel>online_resources</toplevel><creatorcontrib>Cao, Rong</creatorcontrib><creatorcontrib>Bao, Liang</creatorcontrib><creatorcontrib>Wu, Chase</creatorcontrib><creatorcontrib>Zhangsun, Panpan</creatorcontrib><creatorcontrib>Li, Yufei</creatorcontrib><creatorcontrib>Zhang, Zhe</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Cao, Rong</au><au>Bao, Liang</au><au>Wu, Chase</au><au>Zhangsun, Panpan</au><au>Li, Yufei</au><au>Zhang, Zhe</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning</atitle><date>2023-03-28</date><risdate>2023</risdate><abstract>Configuration tuning for large software systems is generally challenging due to the complex configuration space and expensive performance evaluation. Most existing approaches follow a two-phase process, first learning a regression-based performance prediction model on available samples and then searching for the configurations with satisfactory performance using the learned model. Such regression-based models often suffer from the scarcity of samples due to the enormous time and resources required to run a large software system with a specific configuration. Moreover, previous studies have shown that even a highly accurate regression-based model may fail to discern the relative merit between two configurations, whereas performance comparison is actually one fundamental strategy for configuration tuning. To address these issues, this paper proposes CM-CASL, a Comparison-based performance Modeling approach for software systems via Collaborative Active and Semisupervised Learning. CM-CASL learns a classification model that compares the performance of two given configurations, and enhances the samples through a collaborative labeling process by both human experts and classifiers using an integration of active and semisupervised learning. Experimental results demonstrate that CM-CASL outperforms two state-of-the-art performance modeling approaches in terms of both classification accuracy and rank accuracy, and thus provides a better performance model for the subsequent work of configuration tuning.</abstract><doi>10.48550/arxiv.2303.15967</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2303.15967
ispartof
issn
language eng
recordid cdi_arxiv_primary_2303_15967
source arXiv.org
subjects Computer Science - Software Engineering
title CM-CASL: Comparison-based Performance Modeling of Software Systems via Collaborative Active and Semisupervised Learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T20%3A50%3A43IST&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=CM-CASL:%20Comparison-based%20Performance%20Modeling%20of%20Software%20Systems%20via%20Collaborative%20Active%20and%20Semisupervised%20Learning&rft.au=Cao,%20Rong&rft.date=2023-03-28&rft_id=info:doi/10.48550/arxiv.2303.15967&rft_dat=%3Carxiv_GOX%3E2303_15967%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