Automatic CNN Model Partitioning for GPU/FPGA-based Embedded Heterogeneous Accelerators using Geometric Programming - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Journal of Signal Processing Systems Année : 2023

Automatic CNN Model Partitioning for GPU/FPGA-based Embedded Heterogeneous Accelerators using Geometric Programming

Walther Carballo-Hernández
Maxime Pelcat
  • Fonction : Auteur
François Berry

Résumé

Graphics Processing Unit (GPU), dedicated Application Specific Integrated Circuit (ASIC) and Field Programmable Gate Array (FPGA) accelerators are currently platforms of choice for porting Convolutional Neural Networks (CNNs). In this work, an automated Central Processing Unit (CPU)-GPU-FPGA partitioning selection is proposed for a given CNN layer. It is shown that using a Generalized Geometric Programming (GGP) optimization problem formulation, the CPU-GPU-FPGA partitioning problem can be modeled by considering a set of system performance metrics and constraints. Each metric is expressed in a posynomial form depending on CNN hyperparameters and architecture resource models. As for the partitioning method, the state-of-the-art techniques covered are: tiling, grouped convolution and fused-layer. The proposed analytical formalization is then employed to derive a set of objective functions and constraints as a GGP problem. It is demonstrated that it is possible to relax some problem constraints by including a penalization term, and reduce the problem to multiple simpler Geometric Programming (GP) sub-problems. Experimental results targeting an embedded FPGA-GPU platform with CNN layer configurations from state-of-the-art CNN models (AlexNet, VGG16 and ResNet18) show that the simplified problem is solvable in polynomial time with a speed-up gain and energy reduction of around 20% and 15%, respectively, when compared against an arbitrary balanced partitioning. If the models for objective and constraints functions preserve the posynomial form and log-log convexity, it is demonstrated that GGP is an efficient optimization solution to the Design Space Exploration (DSE) problem.
Fichier principal
Vignette du fichier
document.pdf (5.49 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04289176 , version 1 (16-11-2023)

Identifiants

Citer

Walther Carballo-Hernández, Maxime Pelcat, Maxime Pelcat, François Berry. Automatic CNN Model Partitioning for GPU/FPGA-based Embedded Heterogeneous Accelerators using Geometric Programming. Journal of Signal Processing Systems, 2023, 95, pp.1203-1218. ⟨10.1007/s11265-023-01898-0⟩. ⟨hal-04289176⟩
12 Consultations
34 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More