[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117767-en":3,"doc-seo-117767-105":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},117767,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Experimenting with Emerging ARM and RISC-V Systems for Decentralised Machine Learning - Framework and Architecture Evaluation","Decentralised Machine Learning (DML) enables collaborative model training without centralised data sharing, covering approaches such as federated learning and edge inference. Existing tooling often lacks flexibility and portability, limiting experimentation with new processor architectures, non-fully connected topologies, and asynchronous collaboration. A domain-specific language maps DML schemes to the FastFlow parallel runtime, implemented across ARM-v8, RISC-V, and x86-64. Experiments evaluate performance and energy efficiency, and include a public RISC-V port of PyTorch.","Experimenting with Emerging ARM and RISC-V Systems for  \nDecentralised Machine Learning  \narXiv :2302 .07946v1 [ cs .DC] 15 Feb 2023  \nGianluca Mittone  \ngianluca.mittone@unito.it University of Turin Turin, Italy  \nIacopo Colonnelli  \niacopo.colonnelli@unito.it University of Turin Turin, Italy  \nRoberto Esposito  \nroberto.esposito@unito.it University of Turin Turin, Italy  \nMirko Polato  \nmirko.polato@unito.it University of Turin Turin, Italy  \nNicoló Tonci  \n[nicolo.tonci@phd.unipi.it](nicolo.tonci@phd.unipi.it)[ ](nicolo.tonci@phd.unipi.it)University of Pisa Pisa, Italy  \nDoriana Medić  \ndoriana.medic@unito.it University of Turin Turin, Italy  \nEmanuele Parisi  \n[emanuele.parisi@unibo.it](emanuele.parisi@unibo.it)[ ](emanuele.parisi@unibo.it)University of Bologna Bologna, Italy  \nMassimo Torquati  \n[massimo.torquati@unipi.it](massimo.torquati@unipi.it)[ ](massimo.torquati@unipi.it)University of Pisa Pisa, Italy  \nRobert Birke  \nrobert.birke@unito.it University of Turin Turin, Italy  \nAndrea Bartolini  \n[a.bartolini@unibo.it](a.bartolini@unibo.it)[ ](a.bartolini@unibo.it)University of Bologna Bologna, Italy  \nFrancesco Beneventi  \n[francesco.beneventi@unibo.it](francesco.beneventi@unibo.it)[ ](francesco.beneventi@unibo.it)University of Bologna Bologna, Italy  \nLuca Benini  \n[luca.benini@unibo.it](luca.benini@unibo.it)[ ](luca.benini@unibo.it)University of Bologna Bologna, Italy  \nMarco Aldinucci  \nmarco.aldinucci@unito.it University of Turin Turin, Italy  \nABSTRACT  \nDecentralised Machine Learning (DML) enables collaborative machine learning without centralised input data. Federated Learning (FL) and Edge Inference are examples of DML. While tools for DML (especially FL) are starting to flourish, many are not flexible and portable enough to experiment with novel systems (e.g., RISC-V), non-fully connected topologies, and asynchronous collaboration schemes. We overcome these limitations via a domain-specific language allowing to map DML schemes to an underlying middleware, i.e. the FastFlow parallel programming library. We experiment with it by generating different working DML schemes on two emerging architectures (ARM-v8, RISC-V) and the x86-64 platform. We characterise the performance and energy efficiency of the presented schemes and systems. As a byproduct, we introduce a RISC-V porting of the PyTorch framework, the first publicly available to our knowledge.  \nKEYWORDS  \nFederated Learning, Edge Computing, RISC-V, Energy Consumption, Green Computing  \n1 INTRODUCTION  \nRecent years have been characterized by crucial advances in Machine Learning (ML) systems. These advancements have been made  \npossible thanks to the widespread availability of massive and ubiquitous computational resources and copious and distributed data sources. The consequent deployment of ML methods across many industries has generated concerns about data access and movement, such as performance, energy efficiency, security, and privacy [10, 12, 18] . Furthermore, companies consider collected data as competing advantages and are unwilling to share it outside the organization. This results in data being dispersed into isolated islands, and ML practitioners are forbidden from collecting, fusing, and ultimately using the data to improve their systems.  \nDecentralised ML (DML) using Federated Learning (FL) and Edge Inference (EI) tackles the difficulties mentioned above, but practical implementations are not straightforward. The emergence of alternative ISAs to the dominant x86-64, such as ARM-v8 and RISC-V exacerbates the challenges in porting (optimised) libraries and guaranteeing interoperability between heterogeneous systems. Many off-the-shelf frameworks are available, each specifically designed to implement one of a few use-case scenarios. First, while standard FL is based on a master-worker approach, few emerging techniques explore distributed, sparse graphs. Second, to our knowledge, thereis still no FL framework allowing the user to specify a person","cbCaitZt2EuzhR1c","https://ap.wps.com/l/cbCaitZt2EuzhR1c","pdf",5165555,1,9,"English","en",105,"# Introduction\n## Motivation and Limitations of Existing DML Systems\n## Proposed Methodology and Tooling\n## Contributions","[{\"question\":\"Which architectures and learning schemes are evaluated in the experiments?\",\"answer\":\"Experiments generate and test multiple working DML schemes spanning federated learning and edge inference on ARM-v8, RISC-V, and x86-64, assessing both performance and energy efficiency.\"}]","Experimenting with Emerging ARM and RISC-V Systems for Decentralised Machine Learning - Framework and Architecture Evaluation | PDF",1785679458,23,{"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},"experimenting-with-emerging-arm-and-risc-v-systems-for-decentralised-machine-learning-framework-and-architecture-evaluation","",{"@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/experimenting-with-emerging-arm-and-risc-v-systems-for-decentralised-machine-learning-framework-and-architecture-evaluation/117767/",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","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"Which architectures and learning schemes are evaluated in the experiments?","Question",{"text":75,"@type":76},"Experiments generate and test multiple working DML schemes spanning federated learning and edge inference on ARM-v8, RISC-V, and x86-64, assessing both performance and energy efficiency.","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},[85,89,93,97,102,107,112,115,119,122,126],{"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":21,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":98,"slug":129},19,"General","general"]