{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,8,8]],"date-time":"2024-08-08T00:14:33Z","timestamp":1723076073148},"reference-count":39,"publisher":"Association for Computing Machinery (ACM)","issue":"3","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Manag. Data"],"published-print":{"date-parts":[[2024,5,29]]},"abstract":"In response to diverse demands, cloud operators have significantly expanded the array of service offerings, often referred to as Stock Keeping Units (SKUs) available for computing resource configurations. Such diversity has led to increased complexity for customers to choose the appropriate SKU. In the analyzed system, only 43% of the resource capacity was rightly chosen. Although various automated solutions have attempted to resolve this issue, they often rely on the availability of enriched data, such as workload traces, which are unavailable for newly established services. Since these services amass a substantial volume of telemetry from existing users, cloud operators can leverage this information to better understand customer needs and mitigate the risk of over- or under-provisioning. Furthermore, customer satisfaction feedback serves as a crucial resource for continuous learning and improving the recommendation mechanism. In this paper, we present Lorentz, an intelligent SKU recommender for provisioning new compute resources that circumvents the need for workload traces. Lorentz leverages customer profile data to forecast resource capacities for new users based on detailed profiling of existing users. Furthermore, using a continuous learned feedback loop, Lorentz tailors capacity recommendations according to customer performance vs. cost preferences captured through satisfaction signals. Validated using the production data from provisioned VMs supporting Database Platform X, we demonstrate that Lorentz outperforms user selections and existing defaults, reducing slack by >60% without increasing throttling. Evaluated using synthetic data, Lorentz's personalization stage iteratively learns the user preferences over time with high accuracy.<\/jats:p>","DOI":"10.1145\/3654952","type":"journal-article","created":{"date-parts":[[2024,5,30]],"date-time":"2024-05-30T13:44:53Z","timestamp":1717076693000},"page":"1-25","source":"Crossref","is-referenced-by-count":0,"title":["Lorentz: Learned SKU Recommendation Using Profile Data"],"prefix":"10.1145","volume":"2","author":[{"ORCID":"http:\/\/orcid.org\/0009-0000-2356-4587","authenticated-orcid":false,"given":"Nick","family":"Glaze","sequence":"first","affiliation":[{"name":"Microsoft, Cambridge, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0005-3189-7361","authenticated-orcid":false,"given":"Tria","family":"McNeely","sequence":"additional","affiliation":[{"name":"Microsoft, Cambridge, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0005-6857-7505","authenticated-orcid":false,"given":"Yiwen","family":"Zhu","sequence":"additional","affiliation":[{"name":"Microsoft, San Francisco, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0008-8556-5508","authenticated-orcid":false,"given":"Matthew","family":"Gleeson","sequence":"additional","affiliation":[{"name":"Microsoft, Cambridge, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0006-0744-6517","authenticated-orcid":false,"given":"Helen","family":"Serr","sequence":"additional","affiliation":[{"name":"Microsoft, Cambridge, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0006-1209-4632","authenticated-orcid":false,"given":"Rajeev","family":"Bhopi","sequence":"additional","affiliation":[{"name":"Microsoft, Redmond, USA"}]},{"ORCID":"http:\/\/orcid.org\/0009-0007-8534-0889","authenticated-orcid":false,"given":"Subru","family":"Krishnan","sequence":"additional","affiliation":[{"name":"Microsoft, Barcelona, Spain"}]}],"member":"320","published-online":{"date-parts":[[2024,5,30]]},"reference":[{"key":"e_1_2_2_1_1","unstructured":"Amazon.com Inc. 2024. Amazon Web Service. https:\/\/aws.amazon.com\/ Retrieved Jan 4 2024 from"},{"key":"e_1_2_2_2_1","doi-asserted-by":"publisher","DOI":"10.14778\/3554821.3554840"},{"volume-title":"Workload prediction using ARIMA model and its impact on cloud applications' QoS","year":"2014","author":"Calheiros Rodrigo N","key":"e_1_2_2_3_1","unstructured":"Rodrigo N Calheiros, Enayat Masoumi, Rajiv Ranjan, and Rajkumar Buyya. 2014. Workload prediction using ARIMA model and its impact on cloud applications' QoS. IEEE transactions on cloud computing , Vol. 3, 4 (2014), 449--458."},{"volume-title":"Self-Tuning Database Systems: A Decade of Progress. In VLDB '07","year":"2007","author":"Chaudhuri Surajit","key":"e_1_2_2_4_1","unstructured":"Surajit Chaudhuri and Vivek Narasayya. 2007. Self-Tuning Database Systems: A Decade of Progress. In VLDB '07. 3--14."},{"key":"e_1_2_2_5_1","doi-asserted-by":"publisher","DOI":"10.1145\/3132747.3132772"},{"volume-title":"Thayer, et al","year":"2020","author":"Curino Carlo","key":"e_1_2_2_6_1","unstructured":"Carlo Curino, Neha Godwal, Brian Kroth, Sergiy Kuryata, Greg Lapinski, Siqi Liu, Slava Oks, Olga Poppe, Adam Smiechowski, Ed Thayer, et al. 2020. MLOS: An infrastructure for automated software performance engineering. In DEEM. 1--5."},{"key":"e_1_2_2_7_1","doi-asserted-by":"publisher","DOI":"10.1017\/S1351324920000601"},{"key":"e_1_2_2_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2903733"},{"key":"e_1_2_2_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/2644865.2541941"},{"key":"e_1_2_2_10_1","doi-asserted-by":"publisher","DOI":"10.14778\/3137765.3137786"},{"key":"e_1_2_2_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/CNSM.2010.5691343"},{"volume-title":"Proceedings of 9-th International Workshop on Agent and Data Mining Interaction (ADMI-2014)","year":"2014","author":"Gorodetsky Vladimir","key":"e_1_2_2_12_1","unstructured":"Vladimir Gorodetsky, Vladimir Samoylov, and Olga Tushkanova. 2014. Agent-based customer profile learning in 3G recommending systems. In Proceedings of 9-th International Workshop on Agent and Data Mining Interaction (ADMI-2014) Associated with International Conference on Autonomous Agents and Multi-agent Systems (AAMAS-2014), Paris."},{"key":"e_1_2_2_13_1","first-page":"507","article-title":"Why do tree-based models still outperform deep learning on typical tabular data","volume":"35","author":"Grinsztajn L\u00e9o","year":"2022","unstructured":"L\u00e9o Grinsztajn, Edouard Oyallon, and Ga\u00ebl Varoquaux. 2022. Why do tree-based models still outperform deep learning on typical tabular data? Advances in Neural Information Processing Systems , Vol. 35 (2022), 507--520.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_2_14_1","doi-asserted-by":"crossref","unstructured":"Michael Hammer and Arvola Chan. 1976. Index Selection in a Self-Adaptive Data Base Management System. In SIGMOD. 1--8.","DOI":"10.1145\/509383.509385"},{"key":"e_1_2_2_15_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2011.05.027"},{"key":"e_1_2_2_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/APSCC.2010.44"},{"volume-title":"Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems","year":"2017","author":"Ke Guolin","key":"e_1_2_2_17_1","unstructured":"Guolin Ke, Qi Meng, Thomas Finley, Taifeng Wang, Wei Chen, Weidong Ma, Qiwei Ye, and Tie-Yan Liu. 2017. Lightgbm: A highly efficient gradient boosting decision tree. Advances in neural information processing systems , Vol. 30 (2017)."},{"key":"e_1_2_2_18_1","doi-asserted-by":"publisher","DOI":"10.1109\/NOMS.2012.6212065"},{"key":"e_1_2_2_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2015.09.226"},{"volume-title":"Getting political on social network sites: Exploring online political discourse on Facebook. First Monday","year":"2009","author":"Kushin Matthew J","key":"e_1_2_2_20_1","unstructured":"Matthew J Kushin and Kelin Kitchener. 2009. Getting political on social network sites: Exploring online political discourse on Facebook. First Monday (2009)."},{"volume-title":"Python implementation of HALO. https:\/\/github.com\/lotcher\/HALO Retrieved","year":"2024","key":"e_1_2_2_21_1","unstructured":"Lotcher. Bowaer. 2024. Python implementation of HALO. https:\/\/github.com\/lotcher\/HALO Retrieved Jan 4, 2024 from"},{"key":"e_1_2_2_23_1","unstructured":"Oracle. 2021. Oracle Database. https:\/\/www.oracle.com\/database\/. Accessed: 2021--11-03."},{"key":"e_1_2_2_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/1519065.1519068"},{"key":"e_1_2_2_25_1","volume-title":"Proceedings of the international AAAI conference on web and social media","volume":"5","author":"Petrovic Sasa","year":"2011","unstructured":"Sasa Petrovic, Miles Osborne, and Victor Lavrenko. 2011. Rt to win! predicting message propagation in twitter. In Proceedings of the international AAAI conference on web and social media, Vol. 5. 586--589."},{"volume-title":"Seagull: An infrastructure for load prediction and optimized resource allocation. arXiv preprint arXiv:2009.12922","year":"2020","author":"Poppe Olga","key":"e_1_2_2_26_1","unstructured":"Olga Poppe, Tayo Amuneke, Dalitso Banda, Aritra De, Ari Green, Manon Knoertzer, Ehi Nosakhare, Karthik Rajendran, Deepak Shankargouda, Meina Wang, et al. 2020. Seagull: An infrastructure for load prediction and optimized resource allocation. arXiv preprint arXiv:2009.12922 (2020)."},{"key":"e_1_2_2_27_1","doi-asserted-by":"publisher","DOI":"10.14778\/3514061.3514073"},{"key":"e_1_2_2_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/CLOUD.2011.42"},{"key":"e_1_2_2_29_1","doi-asserted-by":"publisher","DOI":"10.1145\/3342195.3387524"},{"key":"e_1_2_2_30_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-24755-2_32"},{"key":"e_1_2_2_31_1","unstructured":"Salesforce Inc. 2023. CRM 101: What is CRM? https:\/\/www.salesforce.com\/crm\/what-is-crm\/ Retrieved Dec 24 2023 from"},{"volume-title":"Uber: Applying Machine Learning to Improve the Customer Pickup Experience. Kellogg School of Management Cases","year":"2019","author":"Sawhney Mohanbir","key":"e_1_2_2_32_1","unstructured":"Mohanbir Sawhney, Birju Shah, Ryan Yu, Evgeny Rubtsov, and Pallavi Goodman. 2019. Uber: Applying Machine Learning to Improve the Customer Pickup Experience. Kellogg School of Management Cases (2019), 1--21."},{"volume-title":"2023 a. GradientBoostingRegression. https:\/\/scikit-learn.org\/stable\/auto_examples\/ensemble\/plot_gradient_boosting_regression.html Retrieved","year":"2023","key":"e_1_2_2_33_1","unstructured":"Scikit-learn. 2023 a. GradientBoostingRegression. https:\/\/scikit-learn.org\/stable\/auto_examples\/ensemble\/plot_gradient_boosting_regression.html Retrieved Oct 4, 2023 from"},{"volume-title":"2023 b. RandomForestRegressor. https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.ensemble.RandomForestRegressor.html Retrieved","year":"2023","key":"e_1_2_2_34_1","unstructured":"Scikit-learn. 2023 b. RandomForestRegressor. https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.ensemble.RandomForestRegressor.html Retrieved Oct 4, 2023 from"},{"key":"e_1_2_2_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3183713.3190650"},{"volume-title":"Amazon Web Service. https:\/\/en.wikipedia.org\/wiki\/Homoscedasticity_and_heteroscedasticity Retrieved","year":"2023","key":"e_1_2_2_36_1","unstructured":"Wikipedia. 2023. Amazon Web Service. https:\/\/en.wikipedia.org\/wiki\/Homoscedasticity_and_heteroscedasticity Retrieved Oct 4, 2023 from"},{"key":"e_1_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3447548.3467190"},{"key":"e_1_2_2_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3457569"},{"key":"e_1_2_2_39_1","doi-asserted-by":"publisher","DOI":"10.1145\/3555041.3589674"},{"key":"e_1_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3412841.3442109"}],"container-title":["Proceedings of the ACM on Management of Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3654952","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T23:01:51Z","timestamp":1723071711000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3654952"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,29]]},"references-count":39,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2024,5,29]]}},"alternative-id":["10.1145\/3654952"],"URL":"https:\/\/doi.org\/10.1145\/3654952","relation":{},"ISSN":["2836-6573"],"issn-type":[{"type":"electronic","value":"2836-6573"}],"subject":[],"published":{"date-parts":[[2024,5,29]]}}}
