Efficient Distributed DNNs in the Mobile-edge-cloud Continuum
This addresses the challenge of efficient distributed learning in heterogeneous mobile-edge-cloud environments, offering a significant performance improvement over existing methods.
The paper tackles the problem of making joint decisions on data selection, model selection, and model-node matching in distributed deep neural networks across the mobile-edge-cloud continuum to minimize energy consumption under learning quality and latency constraints, achieving a competitive ratio of 2(1+ε) and outperforming state-of-the-art solutions by over 50%.
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 $2(1+ε)$, outperforming state-of-the-art solutions by over 50%. Our approach is also validated through a real-world implementation.