[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122045-en":3,"doc-seo-122045-105":30,"detail-sidebar-cat-0-en-105":83},{"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},122045,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","Inferring Network Topology for Distributed Machine Learning Model Training","Distributed machine learning relies increasingly on cloud-based resources for model training, yet many cloud providers withhold information about the underlying network topology for commercial and security reasons. This lack of visibility makes it difficult to place computation modules across different virtual machines to achieve efficient resource utilization. The proposed Flow Tracking (FT) algorithm uses external measurements to infer the internal structure of a general graph. Compared with state-of-the-art methods, FT attains the most accurate topology under four metrics, reaching 100% reconstruction for shortest-path routing. Experiments show topology-driven resource allocation significantly improves training efficiency over random allocation.","Inferring Network Topology for Distributed Machine Learning Model Training  \nby  \nRenjun An  \nB.Eng., Fudan University, 2012  \nM.Sc., Fudan University, 2015  \nA Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of  \nMASTER OF SCIENCE  \nin the Department of Computer Science  \n© Renjun An, 2024 University of Victoria  \nAll rights reserved. This thesis may not be reproduced in whole or in part, by photocopying or other means, without the permission of the author.  \nii  \nInferring Network Topology for Distributed Machine Learning Model Training  \nby  \nRenjun An  \nB.Eng., Fudan University, 2012  \nM.Sc., Fudan University, 2015  \nSupervisory Committee  \nDr. Kui Wu, Supervisor  \n(Department of Computer Science)  \nDr. Jaya Prakash Champati, Departmental Member (Department of Computer Science)  \niii  \nABSTRACT  \nWith the application of distributed machine learning in various industries, there isan increasing demand for model training using cloud computing resources. However, many cloud computing service providers refuse to provide end-users with information about the underlying network topology for commercial and security reasons. Due to this opaqueness, it is challenging to arrange the computation modules in different Virtual Machines (VMs) to achieve the best resource utilization efficiency. To address this problem, we propose an algorithm called Flow Tracking (FT), which uses external measurements to infer the internal structure of a general graph. Compared to the state-of-the-art topology inference algorithms, FT achieves the most accurate topology measured in four different metrics. Notably, FT achieves 100% reconstruction of the underlying topology under the shortest-path routing strategy of the underlying network. Experimentally, resource allocation using the inferred topology improves the model training efficiency significantly compared to random allocation.  \niv  \nContents  \nSupervisory Committee ii  \nAbstract iii  \nContents iv  \nList of Tables vi  \nList of Figures vii  \nAcknowledgements viii  \nDedication ix  \n1 Introduction 1  \n1.1 Background ................................ 1  \n1.1.1 Network Topology Inference ................... 1  \n1.1.2 Distributed Machine Learning .................. 2  \n1.2 Why Topology Matters? ......................... 3  \n1.3 Related Works ............................... 6  \n1.3.1 Topology Inference with Network Tomography ......... 6  \n1.3.2 Topology-aware Cloud Computing ................ 7  \n1.4 Contributions ............................... 7  \n1.5 Organization of Thesis .......................... 8  \n2 System Model and Background 9  \n2.1 Network Model .............................. 9  \n2.2 Measurement Model ........................... 9  \n2.3 Principles in Topology Inference ..................... 11  \nv  \n3 A New Topology Inference Algorithm 14  \n3.1 An Amendment to Category Weight .................. 14  \n3.2 Topology Inference for Cloud Computing ................ 16  \n4 Task Allocation for Model Training 21  \n4.1 Task Allocation Problem ......................... 21  \n4.2 Algorithms to Solve TAP ......................... 22  \n5 Performance Evaluation 26  \n5.1 Experimental Setup ............................ 26  \n5.2 Performance in Topology Inference ................... 27  \n5.3 FT Helps Model Training ........................ 31  \n6 Conclusion and Future Work 36  \n6.1 Conclusion ................................. 36  \n6.2 Future Work ................................ 37  \nBibliography 39  \nvi  \nList of Tables  \nTable 2 . 1 Notations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13  \nvii  \nList of Figures  \nFigure 1.1 Typical data center network topology............... 4  \nFigure 2.1 Example of Inferring Network Topology from Loss Metrics.... 10  \nFigure 3.1 Network topology with non-simple paths............. 16  \nFigure 3.2 An example of topology inference using Flow Tracking (FT) ... 20  \nFigure 5.1 Ground truth topologies...................... 28  \nFigure 5.2 Ho","cbCaiqwECWlyFaWn","https://ap.wps.com/l/cbCaiqwECWlyFaWn","pdf",2939506,1,51,"English","en",105,"# 1 Introduction\n## 1.1 Background\n## 1.2 Why Topology Matters?\n## 1.3 Related Works\n## 1.4 Contributions\n## 1.5 Organization of Thesis\n# 2 System Model and Background\n## 2.1 Network Model\n## 2.2 Measurement Model\n## 2.3 Principles in Topology Inference\n# 3 A New Topology Inference Algorithm\n## 3.1 An Amendment to Category Weight\n## 3.2 Topology Inference for Cloud Computing\n# 4 Task Allocation for Model Training\n## 4.1 Task Allocation Problem\n## 4.2 Algorithms to Solve TAP\n# 5 Performance Evaluation\n## 5.1 Experimental Setup\n## 5.2 Performance in Topology Inference\n## 5.3 FT Helps Model Training\n# 6 Conclusion and Future Work\n## 6.1 Conclusion\n## 6.2 Future Work","[{\"question\":\"How does FT perform compared with existing topology inference methods?\",\"answer\":\"FT achieves the most accurate inferred topology across four different metrics. Under shortest-path routing, it reaches 100% reconstruction of the underlying topology.\"}]","Inferring Network Topology for Distributed Machine Learning Model Training | PDF",1785808542,129,{"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":78,"head_meta":80,"extra_data":82,"updated_unix":28},"inferring-network-topology-for-distributed-machine-learning-model-training","",{"@graph":36,"@context":77},[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/inferring-network-topology-for-distributed-machine-learning-model-training/122045/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"How does FT perform compared with existing topology inference methods?","Question",{"text":75,"@type":76},"FT achieves the most accurate inferred topology across four different metrics. Under shortest-path routing, it reaches 100% reconstruction of the underlying topology.","Answer","https://schema.org",{"og:url":52,"og:type":79,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":81,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]