MétaCan
Menu
← Back to cohort
Record W7141780753

Reconfigurable Programming of Data Plane Communication Networks on FPGA Platforms

2025· other· en· W7141780753 on OpenAlexfundno aff
Parisa Mashreghi-Moghadam

Bibliographic record

VenuePolyPublie (École Polytechnique de Montréal) · 2025
Typeother
Languageen
Field
Topic
Canadian institutionsnot available
FundersNatural Sciences and Engineering Research Council of Canada
KeywordsControl reconfigurationField-programmable gate arrayData transmissionReconfigurable computing
DOInot available

Abstract

fetched live from OpenAlex

RÉSUMÉ: Les réseaux définis par logiciel (Software-Defined Networking — SDN) et le langage P4 ont introduit la programmabilité dans les architectures de plan de données en découplant le comportement de transfert de son implémentation matérielle. Les réseaux prédiffusés programmables (Field-Programmable Gate Array — FPGA) se sont imposés comme une plateforme de choix pour la mise en oeuvre de pipelines de traitement de paquets programmables en P4, combinant reconfigurabilité et parallélisme massif. Cependant, les solutions existantes de compilation de P4 vers FPGA reposent largement sur une spécialisation effectuée au moment de la synthèse, où les architectures d’analyseur et de désassembleur de paquets sont régénérées pour chaque programme P4. Ce modèle statique empêche l’évolution de la logique protocolaire à l’exécution, accroît la latence de déploiement et limite la réutilisation des FPGA dans des environnements dynamiques. Les tentatives visant à améliorer la flexibilité s’appuient souvent sur la reconfiguration de machines à états ou sur une logique de commutation dynamique, mais ces approches entraînent une surcharge de contrôle qui réduit le débit. Par ailleurs, les conceptions basées sur des chemins de données à matrices de commutation larges permettent le réordonnancement des en-têtes, mais entraînent un coût matériel quadratique ainsi qu’une faible évolutivité temporelle sur les FPGA. Cette thèse vise à permettre l’analyse et la reconstruction de paquets programmables à l’exécution sur FPGA, tout en préservant des performances élevées et une utilisation efficace des ressources matérielles. Pour atteindre cet objectif, cette thèse propose un cadre matériellogiciel cohérent permettant de dissocier la sémantique du protocole de l’architecture du chemin de données. Ce cadre introduit deux modes de reconfigurabilité : des architectures à modèles configurables au moment de la synthèse via des génériques VHDL, et des architectures de superposition permettant une reconfiguration en cours d’exécution grâce à des tables de configuration compactes générées automatiquement à partir de descriptions P4. ABSTRACT: Software-Defined Networking (SDN) and the P4 language introduced programmability into data-plane architectures by decoupling forwarding behavior from hardware implementation. Field-Programmable Gate Arrays (FPGAs) have emerged as a compelling platform for implementing P4-programmable packet processing pipelines, combining reconfigurability with high parallelism. However, existing P4-to-FPGA solutions rely heavily on compile-time specialization, where parser and deparser architectures are regenerated for each P4 program. This static design model prevents runtime evolution of protocol logic, increases deployment latency, and limits FPGA reuse in dynamic environments. Attempts to improve flexibility often rely on statemachine reconfiguration or dynamic switching logic, but these approaches introduce control overhead that degrades throughput. Meanwhile, designs based on wide crossbar datapaths support flexible header reordering but incur quadratic hardware cost and exhibit poor timing scalability on FPGAs as datapath widths and protocol complexity grow. This thesis aims to enable runtime-programmable packet parsing and deparsing on FPGAs while preserving high performance and hardware efficiency. To achieve this goal, it introduces a unified hardware–software framework that decouples protocol behavior from datapath structure. The framework provides two levels of reconfigurability: templated architectures, which allow compile-time configuration through VHDL generics, and overlay architectures, which extend this model by enabling runtime reconfiguration through compact memory tables automatically derived from P4 descriptions.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.004
Threshold uncertainty score0.014

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0040.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.025
GPT teacher head0.268
Teacher spread0.243 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venuePolyPublie (École Polytechnique de Montréal)→French-language works237,207→