BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//P4 - Language Consortium - ECPv6.15.20//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:P4 - Language Consortium
X-ORIGINAL-URL:https://p4.org
X-WR-CALDESC:Events for P4 - Language Consortium
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Vancouver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20250309T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20251102T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20260308T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20261101T090000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20270314T100000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20271107T090000
END:STANDARD
END:VTIMEZONE
BEGIN:VTIMEZONE
TZID:UTC
BEGIN:STANDARD
TZOFFSETFROM:+0000
TZOFFSETTO:+0000
TZNAME:UTC
DTSTART:20250101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/Vancouver:20260304T080000
DTEND;TZID=America/Vancouver:20260304T090000
DTSTAMP:20260417T032350
CREATED:20251024T165254Z
LAST-MODIFIED:20260306T175333Z
UID:10000155-1772611200-1772614800@p4.org
SUMMARY:P4 Developer Days - P4Muse (P4 Modularity and Unification for Seamless Extensibility)
DESCRIPTION:Register to attend this P4 Developer Days webinar\, “P4Muse (P4 Modularity and Unification for Seamless Extensibility”\nDate: March 4\, 2026\nTime: 8:00am Pacific \nSlides (PDF) | Watch Recording \nAbstract\n\n\n\nDomain-specific programming languages such as P4 enable flexible and high-performance packet processing in programmable network data planes. However\, most P4 programs remain monolithic\, limiting the development of modular and reusable protocols and libraries. Introducing modularity to P4 has been challenging because existing approaches—such as trans-compilers and virtualization—operate outside the P4 language and compiler\, reducing backward compatibility and extensibility. P4Muse (P4 Modularity and Unification for Seamless Extensibility) addresses this challenge by introducing a compiler-managed approach to modularity within the P4C compiler. Without requiring any new syntax or annotations\, P4Muse introduces new compiler passes for automatic code merging\, enabling modular design\, reuse\, and seamless integration of complex network functionalities. Our results show that P4Muse effectively supports modular P4 program development without altering existing P4 syntax\, providing a robust solution that significantly improves code reusability\, flexibility\, and extensibility while maintaining backward compatibility. \n\n\n\n\nSpeakers\nMohsen Rahmati is a Ph.D. candidate in Computer Engineering at Polytechnique Montréal\, supervised by Professors Yvon Savaria\, François-Raymond Boyer\, and Jean-Pierre David. His research focuses on compiler design for programmable networks\, emphasizing modularity and reusability in the P4 language. He is the creator of P4Muse\, a compiler-managed modularity framework (accepted in IEEE Access\, 2025)\, and P4O2\, an object-oriented-inspired modularity in P4 currently under review at IEEE Access. \n\n  \nFrançois-Raymond Boyer received B.Sc. and Ph.D. degrees in computer science from Université de Montréal\, Montreal\, QC\, Canada\, in 1996 and 2001\, respectively. Since 2001\, he has been with Polytechnique Montréal\, Montréal\, where he is currently a Professor with the Department of Computer and Software Engineering. He has authored or co-authored more than 30 conference and journal papers. His current research interests include microelectronics\, performance optimization\, parallelizing compilers\, digital audio\, and body motion capture. He is a member of Regroupement Stratégique en Microélectronique du Québec\, Groupe de Recherche en Microélectronique et Microsystèmes\, and Observatoire Interdisciplinaire de Création et de Recherche en Musique.
URL:https://p4.org/event/p4-developer-days-p4muse-p4-modularity-and-unification-for-seamless-extensibility/
CATEGORIES:Events
ATTACH;FMTTYPE=image/png:https://p4.org/wp-content/uploads/sites/53/2025/10/Dev-Day-March-4.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=UTC:20260319T080000
DTEND;TZID=UTC:20260319T090000
DTSTAMP:20260417T032350
CREATED:20251117T170038Z
LAST-MODIFIED:20260320T115620Z
UID:10000156-1773907200-1773910800@p4.org
SUMMARY:P4 Developer Days - SpliDT: Partitioned Decision Trees for Scalable Stateful Inference at Line Rate
DESCRIPTION:Register to attend this P4 Developer Days webinar\, “SpliDT: Partitioned Decision Trees for Scalable Stateful Inference at Line Rate”\nMarch 19 at 11 am ET/4 pm CET \nSlides (PDF) || Watch Recording \n\n\n\nAbstract\nMachine learning is increasingly used in programmable data planes\, such as switches and smartNICs\, to enable real-time traffic analysis and security monitoring at line rate. Decision trees (DTs) are particularly well-suited for these tasks due to their interpretability and compatibility with the Reconfigurable Match-Action Table (RMT) architecture. However\, current DT implementations require collecting all features upfront\, which limits scalability and accuracy due to constrained data plane resources. This paper introduces SpliDT\, a scalable framework that reimagines DT deployment as a partitioned inference problem over a sliding window of packets. \nBy dividing inference into sequential subtrees—each using its own set of top-k features—SpliDT supports more stateful features without exceeding hardware limits. An in-band control channel manages transitions between subtrees and reuses match keys and registers across partitions\, implemented using P4 for packet recirculation and control-plane coordination. This design allows physical resources to be shared efficiently while maintaining line-rate processing. To maximize accuracy and scalability\, SpliDT employs a custom training and design-space-exploration (DSE) workflow that jointly optimizes feature allocation\, tree depth\, and partitioning. Evaluations show that SpliDT supports up to 5x more features\, scales to millions of flows\, and outperforms baselines\, with low overhead and minimal time-to-detection (TTD). \n\n\nSpeaker\nMurayyiam Parvez is a Ph.D. candidate in the Department of Computer Science at Purdue University\, advised by Professor Muhammad Shahbaz and Professor Sonia Fahmy. Her research focuses on designing the next generation of hybrid software–hardware abstractions and architectures for emerging network security and machine learning applications\, with an emphasis on leveraging programmable switches to enable high-performance\, in-network intelligence. \n\nRegister to attend this webinar!
URL:https://p4.org/event/p4-developer-days-splidt-partitioned-decision-trees-for-scalable-stateful-inference-at-line-rate/
CATEGORIES:Events
ATTACH;FMTTYPE=image/png:https://p4.org/wp-content/uploads/sites/53/2025/11/Dev-Day-March-19.png
END:VEVENT
END:VCALENDAR