Efficient Distributed DNNs in the Mobile-Edge-Cloud Continuum

• Published in: IEEE/ACM transactions on networking Vol. 31; no. 4; pp. 1 - 15
• Main Authors: Malandrino, Francesco, Chiasserini, Carla Fabiana, di Giacomo, Giuseppe
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: 5G and beyond networks; Cognitive tasks; Controllers; Data sources; Decisions; Distributed machine learning; Energy consumption; Graph representations; Graphical representations; Machine learning; mobile-edge-cloud continuum; Nodes; Polynomials
• ISSN: 1063-6692; 1558-2566
• DOI: 10.1109/TNET.2022.3222640

In the mobile-edge-cloud continuum, a plethora of heterogeneous data sources and computation-capable nodes are available. Such nodes can cooperate to perform a distributed learning task, aided by a learning controller (often located at the network edge). The controller is required to make decisions concerning (i) data selection, i.e., which data sources to use; (ii) model selection, i.e., which machine learning model to adopt, and (iii) matching between the layers of the model and the available physical nodes. All these decisions influence each other, to a significant extent and often in counter-intuitive ways. In this paper, we formulate a problem addressing all of the above aspects and present a solution concept called RightTrain, aiming at making the aforementioned decisions in a joint manner, minimizing energy consumption subject to learning quality and latency constraints. RightTrain leverages an expanded-graph representation of the system and a delay-aware Steiner tree to obtain a provably near-optimal solution while keeping the time complexity low. Specifically, it runs in polynomial time and its decisions exhibit a competitive ratio of <inline-formula> <tex-math notation="LaTeX">2(1+\epsilon)</tex-math> </inline-formula>, outperforming state-of-the-art solutions by over 50%. Our approach is also validated through a real-world implementation.