Minimalistic Predictions to Schedule Jobs with Online Precedence Constraints

TL;DR

This paper explores non-clairvoyant online scheduling with precedence constraints using predictions, providing bounds and insights into how additional information can improve scheduling performance.

Contribution

It introduces novel prediction models for scheduling with precedence constraints and offers bounds and algorithms that leverage these predictions to improve performance.

Findings

Lower bounds and upper bounds for various precedence topologies.

Structured overview of how predictions influence scheduling efficiency.

Improved bounds on traditional competitive ratios.

Abstract

We consider non-clairvoyant scheduling with online precedence constraints, where an algorithm is oblivious to any job dependencies and learns about a job only if all of its predecessors have been completed. Given strong impossibility results in classical competitive analysis, we investigate the problem in a learning-augmented setting, where an algorithm has access to predictions without any quality guarantee. We discuss different prediction models: novel problem-specific models as well as general ones, which have been proposed in previous works. We present lower bounds and algorithmic upper bounds for different precedence topologies, and thereby give a structured overview on which and how additional (possibly erroneous) information helps for designing better algorithms. Along the way, we also improve bounds on traditional competitive ratios for existing algorithms.