Table of Contents
Fetching ...

Beyond 2-Edge-Connectivity: Algorithms and Impossibility for Content-Oblivious Leader Election

Yi-Jun Chang, Lyuting Chen, Haoran Zhou

TL;DR

This work characterizes the feasibility of terminating leader election in the content-oblivious model on trees with topology knowledge, showing that termination is possible if and only if the tree is not symmetric about any edge. It provides two main algorithms: a diameter-aware approach for even-diameter trees achieving O(nr) pulses and a topology-aware approach for general asymmetric trees achieving O(n^2) pulses, while proving impossibility for edge-symmetric graphs even with topology knowledge. The results establish that topology knowledge is essential in non-2-edge-connected graphs and offer a stabilizing, uniform leader election method that does not rely on prior network-size knowledge. Overall, the paper advances understanding of content-oblivious computation beyond 2-edge-connectivity and lays out open questions about extending these techniques to broader graph classes and improving efficiency.

Abstract

The content-oblivious model, introduced by Censor-Hillel, Cohen, Gelles, and Sel (PODC 2022; Distributed Computing 2023), captures an extremely weak form of communication where nodes can only send asynchronous, content-less pulses. Censor-Hillel, Cohen, Gelles, and Sel showed that no non-constant function $f(x,y)$ can be computed correctly by two parties using content-oblivious communication over a single edge, where one party holds $x$ and the other holds $y$. This seemingly ruled out many natural graph problems on non-2-edge-connected graphs. In this work, we show that, with the knowledge of network topology $G$, leader election is possible in a wide range of graphs. Impossibility: Graphs symmetric about an edge admit no randomized terminating leader election algorithm, even when nodes have unique identifiers and full knowledge of $G$. Leader election algorithms: Trees that are not symmetric about any edge admit a quiescently terminating leader election algorithm with topology knowledge, even in anonymous networks, using $O(n^2)$ messages, where $n$ is the number of nodes. Moreover, even-diameter trees admit a terminating leader election given only the knowledge of the network diameter $D = 2r$, with message complexity $O(nr)$. Necessity of topology knowledge: In the family of graphs $\mathcal{G} = \{P_3, P_5\}$, both the 3-path $P_3$ and the 5-path $P_5$ admit a quiescently terminating leader election if nodes know the topology exactly. However, if nodes only know that the underlying topology belongs to $\mathcal{G}$, then terminating leader election is impossible.

Beyond 2-Edge-Connectivity: Algorithms and Impossibility for Content-Oblivious Leader Election

TL;DR

This work characterizes the feasibility of terminating leader election in the content-oblivious model on trees with topology knowledge, showing that termination is possible if and only if the tree is not symmetric about any edge. It provides two main algorithms: a diameter-aware approach for even-diameter trees achieving O(nr) pulses and a topology-aware approach for general asymmetric trees achieving O(n^2) pulses, while proving impossibility for edge-symmetric graphs even with topology knowledge. The results establish that topology knowledge is essential in non-2-edge-connected graphs and offer a stabilizing, uniform leader election method that does not rely on prior network-size knowledge. Overall, the paper advances understanding of content-oblivious computation beyond 2-edge-connectivity and lays out open questions about extending these techniques to broader graph classes and improving efficiency.

Abstract

The content-oblivious model, introduced by Censor-Hillel, Cohen, Gelles, and Sel (PODC 2022; Distributed Computing 2023), captures an extremely weak form of communication where nodes can only send asynchronous, content-less pulses. Censor-Hillel, Cohen, Gelles, and Sel showed that no non-constant function can be computed correctly by two parties using content-oblivious communication over a single edge, where one party holds and the other holds . This seemingly ruled out many natural graph problems on non-2-edge-connected graphs. In this work, we show that, with the knowledge of network topology , leader election is possible in a wide range of graphs. Impossibility: Graphs symmetric about an edge admit no randomized terminating leader election algorithm, even when nodes have unique identifiers and full knowledge of . Leader election algorithms: Trees that are not symmetric about any edge admit a quiescently terminating leader election algorithm with topology knowledge, even in anonymous networks, using messages, where is the number of nodes. Moreover, even-diameter trees admit a terminating leader election given only the knowledge of the network diameter , with message complexity . Necessity of topology knowledge: In the family of graphs , both the 3-path and the 5-path admit a quiescently terminating leader election if nodes know the topology exactly. However, if nodes only know that the underlying topology belongs to , then terminating leader election is impossible.

Paper Structure

This paper contains 50 sections, 39 theorems, 22 equations, 2 algorithms.

Key Result

Theorem 1.2

There exists a terminating, anonymous content-oblivious leader election algorithm with message complexity $O(nr)$ for a tree with even diameter $D = 2r$ and $n$ nodes, provided that each node knows the diameter $D$ of the tree a priori.

Theorems & Definitions (74)

  • Definition 1.1: Edge symmetry
  • Theorem 1.2: Leader election on even-diameter trees
  • Theorem 1.3: Leader election on general asymmetric trees
  • Theorem 1.4: Edge symmetry implies impossibility, randomized
  • Corollary 1.4: Edge symmetry implies impossibility, deterministic
  • Theorem 1.5: Necessity of topology knowledge
  • Theorem 1.6: Stabilizing leader election
  • Theorem 3.1: Leader election on even-diameter trees
  • proof
  • Definition 3.2: Parent and children
  • ...and 64 more