[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119604-en":3,"doc-seo-119604-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119604,1099514068365,"Aurelia","https://ap-avatar.wpscdn.com/avatar/10000253d8d9f28188e?_k=1776742907772140068",8,"Research & Report","Machine Learning Methodologies for Supporting HPC Systems Operations - Thesis","This thesis advances machine learning research while addressing real-life engineering needs in high-performance computing (HPC) system monitoring and management. It develops an operational data analytics (ODA) framework that spans data collection through processing and visualization, targeting open-ended data exploration, unsupervised anomaly detection, and long-term anomaly prediction. The framework defines adoption stages that enable subsequent components. It introduces DEM for label-free exploration and RUAD for temporal, recurrent unsupervised anomaly detection, evaluated on the Marconi 100 system.","DOTTORATO DI RICERCA IN  \nDATA SCIENCE AND COMPUTATION  \nCiclo 36  \nSettore Concorsuale: 09/H1-SISTEMI DI ELABORAZIONE DELLE INFORMAZIONI  \nSettore Scientifico Disciplinare: ING-INF/05-SISTEMI DI ELABORAZIONE DELLE INFORMAZIONI  \nMACHINE LEARNING METHODOLOGIES FOR SUPPORTING HPC SYSTEMS  \nOPERATIONS  \nPresentata da: Martin Molan  \nCoordinatore Dottorato  \nDaniele Bonacorsi  \nSupervisore  \nAndrea Bartolini  \nCo-supervisore  \nLuca Benini  \nEsame finale anno 2025  \niii  \nDeclaration of Authorship  \nI, Martin MOLAN, declare that this thesis titled,  \nMachine Learning methodologies to support HPC systems operations and the work presented in it are my own. I confirm that:  \n• This work was done wholly or mainly while in candidature for a research degree at this University.  \n• Where any part of this thesis has previously been submitted for a degree or any other qualification at this University or any other institution, this has been clearly stated.  \n• Where I have consulted the published work of others, this is always clearly attributed.  \n• Where I have quoted from the work of others, the source is always given. With the exception of such quotations, this thesis is entirely my own work.  \n• I have acknowledged all main sources of help.  \n• Where the thesis is based on work done by myself jointly with others, I have made clear exactly what was done by others and what I have contributed myself.  \nSigned:  \nDate:  \nv  \n“But I made myself ready to endure, and, aided by your words, I made my way along the rocky slope, behind my guide.”  \nInferno, Canto XXIV, Dante Alighieri  \nvii  \nALMA MATER STUDIORUM  \nUNIVERSITY OF BOLOGNA  \nAbstract  \nFaculty of Engineering  \nDepartment of Electrical, Electronic, and Information Engineering PhD in Data Science and Computation  \nMachine Learning Methodologies for Supporting HPC Systems Operations  \nby Martin MOLAN  \nviii  \nThe work presented in this thesis has been driven by two equal and interconnected principles: performing fundamental machine learning research and solving reallife engineering problems. The specific target domain is high-performance computing (HPC) systems and the specific challenges that come with adapting machine learning methodologies for their monitoring and management. However, models, findings, and methodologies developed for and motivated by this specific set of requirements have applications far beyond the original domain.  \nThe increasing size and complexity of modern HPC systems necessitate introducing advanced data collection, monitoring, and machine learning methodologies that support their management and operations. In literature, this collection of methodologies from data collection to data processing and visualization is called operational data analytics (ODA) for HPC systems. The thesis presents and discusses the comprehensive ODA framework comprising multiple models that address some of the most pressing open problems in the field: open-ended data exploration, unsupervised anomaly detection, and long-term anomaly prediction.  \nThe ODA framework first establishes the continuum of the machine learning model adoption in the HPC systems, with each part of the framework addressing aspecific stage with its unique requirements and previously unanswered questions. Depending on the level of adoption of operational data analytic methodologies, HPC systems can adopt one, some, or all parts of the framework. The stages of the framework support each other; however, each stage, besides solving its primary objective, enables the adoption of the next one.  \nThe first part of the comprehensive ODA framework is the methodology to perform open-ended data exploration and analysis, named the DEM (Data exploration model) . DEM is the foundation of the comprehensive ODA framework as it requires no structured or labeled data and can thus be deployed as the first machine-learning model adopted by the HPC system. DEM provides insights into the operation of the nodes and allows the HPC system administ","cbCailaWILBk5Ump","https://ap.wps.com/l/cbCailaWILBk5Ump","pdf",14534729,1,192,"English","en",105,"# Abstract\n## Operational Data Analytics (ODA) Framework\n## DEM: Data Exploration Model\n## RUAD: Recurrent Unsupervised Anomaly Detection\n## GRAAFE: Long-Term Anomaly Prediction","[{\"question\":\"What problem does the thesis focus on in HPC systems operations?\",\"answer\":\"It focuses on adapting machine learning methodologies for monitoring and management of high-performance computing systems under the challenges posed by their scale and complexity.\"},{\"question\":\"What is the ODA framework and what does it cover?\",\"answer\":\"The operational data analytics (ODA) framework covers multiple stages from data collection to processing and visualization, supporting open-ended exploration, unsupervised anomaly detection, and long-term anomaly prediction.\"},{\"question\":\"How do DEM and RUAD fit into the framework?\",\"answer\":\"DEM provides label-free open-ended data exploration to surface relevant metrics for further analysis. RUAD delivers recurrent unsupervised anomaly detection by leveraging temporal dependencies in HPC data.\"}]","Machine Learning Methodologies for Supporting HPC Systems Operations - Thesis | PDF",1785725239,484,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-methodologies-for-supporting-hpc-systems-operations-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-methodologies-for-supporting-hpc-systems-operations-thesis/119604/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the thesis focus on in HPC systems operations?","Question",{"text":75,"@type":76},"It focuses on adapting machine learning methodologies for monitoring and management of high-performance computing systems under the challenges posed by their scale and complexity.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the ODA framework and what does it cover?",{"text":80,"@type":76},"The operational data analytics (ODA) framework covers multiple stages from data collection to processing and visualization, supporting open-ended exploration, unsupervised anomaly detection, and long-term anomaly prediction.",{"name":82,"@type":73,"acceptedAnswer":83},"How do DEM and RUAD fit into the framework?",{"text":84,"@type":76},"DEM provides label-free open-ended data exploration to surface relevant metrics for further analysis. RUAD delivers recurrent unsupervised anomaly detection by leveraging temporal dependencies in HPC data.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]