ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2007.14898
28
34

Deterministic Distributed Expander Decomposition and Routing with Applications in Distributed Derandomization

29 July 2020
Yi-Jun Chang
Thatchaphol Saranurak
ArXiv (abs)PDFHTML
Abstract

There is a recent exciting line of work in distributed graph algorithms in the CONGEST\mathsf{CONGEST}CONGEST model that exploit expanders. All these algorithms so far are based on two tools: expander decomposition and expander routing. An (ϵ,ϕ)(\epsilon,\phi)(ϵ,ϕ)-expander decomposition removes ϵ\epsilonϵ-fraction of the edges so that the remaining connected components have conductance at least ϕ\phiϕ, i.e., they are ϕ\phiϕ-expanders, and expander routing allows each vertex vvv in a ϕ\phiϕ-expander to very quickly exchange deg(v)\text{deg}(v)deg(v) messages with any other vertices, not just its local neighbors. In this paper, we give the first efficient deterministic distributed algorithms for both tools. We show that an (ϵ,ϕ)(\epsilon,\phi)(ϵ,ϕ)-expander decomposition can be deterministically computed in poly(ϵ−1)no(1)\text{poly}(\epsilon^{-1}) n^{o(1)}poly(ϵ−1)no(1) rounds for ϕ=poly(ϵ)n−o(1)\phi = \text{poly}(\epsilon) n^{-o(1)}ϕ=poly(ϵ)n−o(1), and that expander routing can be performed deterministically in poly(ϕ−1)no(1)\text{poly}(\phi^{-1})n^{o(1)}poly(ϕ−1)no(1) rounds. Both results match previous bounds of randomized algorithms by [Chang and Saranurak, PODC 2019] and [Ghaffari, Kuhn, and Su, PODC 2017] up to subpolynomial factors. Consequently, we derandomize existing distributed algorithms that exploit expanders. We show that a minimum spanning tree on no(1)n^{o(1)}no(1)-expanders can be constructed deterministically in no(1)n^{o(1)}no(1) rounds, and triangle detection and enumeration on general graphs can be solved deterministically in O(n0.58)O(n^{0.58})O(n0.58) and n2/3+o(1)n^{2/3 + o(1)}n2/3+o(1) rounds, respectively. We also give the first polylogarithmic-round randomized algorithm for constructing an (ϵ,ϕ)(\epsilon,\phi)(ϵ,ϕ)-expander decomposition in poly(ϵ−1,log⁡n)\text{poly}(\epsilon^{-1}, \log n)poly(ϵ−1,logn) rounds for ϕ=1/poly(ϵ−1,log⁡n)\phi = 1 / \text{poly}(\epsilon^{-1}, \log n)ϕ=1/poly(ϵ−1,logn). The previous algorithm by [Chang and Saranurak, PODC 2019] needs nΩ(1)n^{\Omega(1)}nΩ(1) rounds for any ϕ≥1/polylog⁡n\phi\ge 1/\text{poly}\log nϕ≥1/polylogn.

View on arXiv
Comments on this paper