[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124906-en":3,"doc-seo-124906-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"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},124906,4398048950312,"Violet","https://ap-avatar.wpscdn.com/avatar/400002538284de19e3c?_k=1778320343897328908",8,"Research & Report","Machine learning enhanced code optimization for high-level synthesis (ML-ECOHS)","While Field-Programmable Gate Arrays (FPGAs) are used across data centers, cloud, and edge platforms, their performance and flexibility often remain limited by hardware design expertise, leaving most computation fixed in CPUs, GPUs, and ASICs. This dissertation advances High Level Synthesis (HLS) by leveraging machine learning to generate FPGA configurations from unmodified high-level language programs. The work addresses the serial execution model mismatch by applying transformation-driven parallelization and then using ML to select promising optimizations and ordering.","Boston University  \nOpenBU [http://open. bu.edu](http://open. bu.edu)  \nTheses & Dissertations Boston University Theses & Dissertations  \n2024  \nMachine learning enhanced code optimization for high-level synthesis (ML-ECOHS)  \n[https://hdl.handle.net/2144/48859](https://hdl.handle.net/2144/48859)[ ](https://hdl.handle.net/2144/48859)Boston University  \nBOSTON UNIVERSITY  \nCOLLEGE OF ENGINEERING  \nDissertation  \nMACHINE LEARNING ENHANCED CODE OPTIMIZATION FOR HIGH-LEVEL SYNTHESIS (ML-ECOHS)  \nby  \nROBERT P. MUNAFO  \nA.B., Dartmouth College, 1986  \nM.S., Boston University, 2018  \nSubmitted in partial fulfillment of the requirements for the degree of  \nDoctor of Philosophy  \n© 2024 by  \nROBERT P. MUNAFO All rights reserved  \nApproved by  \nFirst Reader  \nMartin C. Herbordt, Ph.D.  \nProfessor of Electrical and Computer Engineering  \nSecond Reader  \nRichard Brower, Ph.D.  \nProfessor of Electrical and Computer Engineering  \nThird Reader  \nTali Moreshet, Ph.D.  \nMaster Lecturer & Research Assistant Professor of Electrical and Computer Engineering  \nFourth Reader  \nThomas D. VanCourt, Ph.D. Software Engineer Akamai Technologies  \nOther experimenters will repeat your experiment and find out  \nwhether you were wrong or right. Nature's phenomena will agree or they'll disagree with your theory . . .  \n—Richard P. Feynman  \nAcknowledgments  \nI wish to acknowledge my thesis advisor Prof. Martin Herbordt for the many meetings to discuss and clarify ideas, and for extensive patience.  \nFor their technical advice and material help, I gratefully acknowledge our projectspecific advisors Uli Drepper, Sanjay Arora, and Ahmed Sanaullah.  \nIn addition I acknowledge my thesis committee members—Profs. Richard Brower and Tali Moreshet, and Dr. Tom VanCourt—for their time and effort in reading, and responding with pokes and prods to keep me going, to encourage new directions, and to heighten clarity of communication.  \nThe other members of our CAAD lab, most notably Hafsah Shahzad, have answered many questions small and large and helped explain many details of their work that relate to this work.  \nIt also goes nearly without saying that I have friends and family, without whose help and support I would not have gotten this far. Thank you, particularly to the friends and colleagues who encouraged me to go back to school in the first place.  \n—Robert Munafo  \nMACHINE LEARNING ENHANCED CODE OPTIMIZATION FOR HIGH-LEVEL SYNTHESIS (ML-ECOHS)  \nROBERT P. MUNAFO  \nBoston University, College of Engineering, 2024  \nMajor Professor: Martin Herbordt, PhD  \nProfessor of Electrical and Computer Engineering  \nABSTRACT  \nWhile Field-Programmable Gate Arrays (FPGAs) exist in many design configurations throughout the data center, cloud, and edge, the promise of performance and flexibility offered by the FPGA often remains unrealized for lack of hardware design expertise, with most computation remaining in fixed hardware such as CPUs, GPUs, and ASICs e.g. tensor processors. Identifying programmability as a barrier to FPGA usage, we seek to augment High Level Synthesis (HLS) design flows with machine learning. The overall goal of this dissertation is to advance the art of using unmodified high-level language (HLL) programs to create FPGA configurations that are performant, programmable, and portable.  \nThe problems in using HLL code to program FPGAs arise from the serial execution model of the target application codes, in particular, the mismatch between that model and the arbitrary data flow model of the target hardware. However, a variety of code transformation techniques, tedious to perform by a human but readily and effortlessly done by CPU compilers, allow many compute-intensive operations to be transformed into a form that is highly or massively parallel. A challenge then exists in selecting the best set of optimizations and an order in which to perform them, a choice among staggeringly many options. Brute-force and automated orthogonal search techniques have failed to produce s","cbCairekslcTulEl","https://ap.wps.com/l/cbCairekslcTulEl","pdf",3248009,1,243,"English","en",105,"# Introduction\n## Motivation and Problem Statement\n## Statement of Thesis\n## Practical PPP via HLS with AI Assistance\n## Contributions\n## Outline\n# Background\n## Target Hardware Environments\n## HPC Optimization and Automation Thereof\n## Non-HLL-Modellable Hardware Accelerators\n## Field-Programmable Gate Arrays","[{\"question\":\"What problem does ML-ECOHS target in using FPGAs with high-level languages?\",\"answer\":\"It targets the difficulty of generating performant FPGA designs from unmodified high-level language code, mainly caused by the mismatch between serial execution models of target applications and the arbitrary dataflow model of hardware.\"},{\"question\":\"How does the dissertation incorporate machine learning into HLS workflows?\",\"answer\":\"It evaluates ML models to recommend code transformation optimizations and their ordering, using compilation results as feedback. It also develops methods for preparing HLL programs as model inputs and applying model outputs to a compilation system.\"},{\"question\":\"Why are brute-force or automated orthogonal search methods inadequate here?\",\"answer\":\"The choice among an extremely large number of optimization options creates a search space where brute-force and automated orthogonal search cannot find practical solutions within feasible limits.\"}]","Machine learning enhanced code optimization for high-level synthesis (ML-ECOHS) | PDF",1785895328,612,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-enhanced-code-optimization-for-high-level-synthesis-ml-ecohs","",{"@graph":36,"@context":86},[37,54,69],{"@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-enhanced-code-optimization-for-high-level-synthesis-ml-ecohs/124906/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-06","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does ML-ECOHS target in using FPGAs with high-level languages?","Question",{"text":76,"@type":77},"It targets the difficulty of generating performant FPGA designs from unmodified high-level language code, mainly caused by the mismatch between serial execution models of target applications and the arbitrary dataflow model of hardware.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the dissertation incorporate machine learning into HLS workflows?",{"text":81,"@type":77},"It evaluates ML models to recommend code transformation optimizations and their ordering, using compilation results as feedback. It also develops methods for preparing HLL programs as model inputs and applying model outputs to a compilation system.",{"name":83,"@type":74,"acceptedAnswer":84},"Why are brute-force or automated orthogonal search methods inadequate here?",{"text":85,"@type":77},"The choice among an extremely large number of optimization options creates a search space where brute-force and automated orthogonal search cannot find practical solutions within feasible limits.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]