Lower Bounds for the Fair Resource Allocation Problem - Archive ouverte HAL Access content directly
Conference Papers Year :

Lower Bounds for the Fair Resource Allocation Problem

(1, 2) , (1) , (2) , (2)
1
2

Abstract

The α-fair resource allocation problem has received remarkable attention and has been studied in numerous application fields. Several algorithms have been proposed in the context of α-fair resource sharing to distributively compute its value. However, little work has been done on its structural properties. In this work, we present a lower bound for the optimal solution of the weighted α-fair resource allocation problem and compare it with existing propositions in the literature. Our derivations rely on a localization property verified by optimization problems with separable objective that permit one to better exploit their local structures. We give a local version of the well-known midpoint domination axiom used to axiomatically build the Nash Bargaining Solution (or proportionally fair resource allocation problem). Moreover, we show how our lower bound can improve the performances of a distributed algorithm based on the Alternating Directions Method of Multipliers (ADMM). The evaluation of the algorithm shows that our lower bound can considerably reduce its convergence time up to two orders of magnitude compared to when the bound is not used at all or is simply looser.
Fichier principal
Vignette du fichier
allybokus2017lower(1).pdf (4.21 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01706142 , version 1 (10-02-2018)

Identifiers

  • HAL Id : hal-01706142 , version 1

Cite

Zaid Allybokus, Konstantin Avrachenkov, Jérémie Leguay, Lorenzo Maggi. Lower Bounds for the Fair Resource Allocation Problem. IFIP Performance 2017 - 35th International Symposium on Computer Performance, Modeling, Measurements and Evaluation, Nov 2017, New York, United States. pp.1-7. ⟨hal-01706142⟩
64 View
105 Download

Share

Gmail Facebook Twitter LinkedIn More