A Survey of Stability Results for Redundancy Systems

TL;DR

This survey reviews recent stability results for redundancy systems with cancel-on-completion, highlighting their advantages, limitations, and open questions in realistic multi-server models with diverse service-time distributions.

Contribution

It provides the first comprehensive benchmark of stability in systems with cancel-on-completion redundancy, summarizing recent findings and identifying open research questions.

Findings

Redundancy can improve system stability under certain conditions.

Stability depends on service-time distributions and correlation structures.

Open problems remain in modeling realistic scenarios.

Abstract

Redundancy mechanisms consist in sending several copies of a same job to a subset of servers. It constitutes one of the most promising ways to exploit diversity in multiservers applications. However, its pros and cons are still not sufficiently understood in the context of realistic models with generic statistical properties of service-times distributions and correlation structures of copies. We aim at giving a survey of recent results concerning the stability-arguably the first benchmark of performance-of systems with cancel-oncompletion redundancy. We also point out open questions and conjectures.