no code implementations • 25 Sep 2023 • Amine Barrak, Mayssa Jaziri, Ranim Trabelsi, Fehmi Jaafar, Fabio Petrillo
The advent of serverless computing has ushered in notable advancements in distributed machine learning, particularly within parameter server-based architectures.
1 code implementation • 25 Sep 2023 • Amine Barral, Ranim Trabelsi, Fehmi Jaafar, Fabio Petrillo
In this paper, we introduce a novel architecture that combines serverless computing with P2P networks for distributed training and present a method for efficient parallel gradient computation under resource constraints.
no code implementations • 27 Feb 2023 • Amine Barrak, Fabio Petrillo, Fehmi Jaafar
However, the parameter server architecture may have limitations in terms of fault tolerance, including a single point of failure and complex recovery processes.