Forwarding Packets Greedily

TL;DR

This paper investigates online packet forwarding in line networks, showing a greedy algorithm's optimal competitiveness for certain cases and establishing a lower bound for all algorithms, addressing an open problem from 2014.

Contribution

It introduces a greedy algorithm with proven competitive ratio for specific forwarding scenarios and provides a lower bound, advancing understanding of online packet forwarding.

Findings

Greedy algorithm achieves ratio of 2-2^{1-k} for packets needing 1 or 2 routers.

Lower bound of 4/3 for randomized algorithms in this problem.

Progress on an open problem from Antoniadis et al. (2014).

Abstract

We consider the problem of forwarding packets arriving online with their destinations in a line network. In each time step, each router can forward one packet along the edge to its right. Each packet that is forwarded arrives at the next router one time step later. Packets are forwarded until they reach their destination. The flow time of a packet is the difference between its release time and the time of its arrival at its destination. The goal is to minimize the maximum flow time. This problem was introduced by Antoniadis et al.~in 2014. They propose a collection of natural algorithms and prove for one, and claim for others, that none of them are $O(1)$-competitive. It was posed as an open problem whether such an algorithm exists. We make the first progress on answering this question. We consider the special case where each packet needs to be forwarded by exactly one or two routers. We show that a greedy algorithm, which was not previously considered for this problem, achieves a competitive ratio of exactly $2-2^{1-k}$, where $k$ is the number of active routers in the network. We also give a general lower bound of $4/3$, even for randomized algorithms.