Fair Ordering

2025-10-15

AI Summary

In distributed systems, clock asynchrony impedes fair and deterministic total ordering of events. Method: This paper proposes a probabilistic event serialization framework that replaces the deterministic “happened-before” relation with a novel “likely-happened-before” probabilistic ordering. It models per-node clock skews as time-varying probability distributions and performs Bayesian comparison of noisy timestamps to infer the most probable causal order. Crucially, it requires no strong clock synchronization assumptions and operates online. Contribution/Results: The approach enables fine-grained, statistically fair total order generation without global consensus or physical clock alignment. Experiments demonstrate accurate estimation of relative occurrence probabilities between events. The method provides a provably sound, scalable foundation—both theoretically and empirically—for achieving system-wide fairness in asynchronous environments.

Technology Category

• Reasoning under Uncertainty: Relational Probabilistic Models
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Distributed Search

Application Category

• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

A growing class of applications demands emph{fair ordering/sequencing} of events which ensures that events generated earlier by one client are processed before later events from other clients. However, achieving such sequencing is fundamentally challenging due to the inherent limitations of clock synchronization. We advocate for an approach that embraces, rather than eliminates, clock variability. Instead of attempting to remove error from a timestamp, Tommy, our proposed system, leverages a statistical model to compare two noisy timestamps probabilistically by learning per-clock offset distributions. Our preliminary statistical model computes the probability that one event precedes another w.r.t. the wall-clock time without access to the wall-clock. This serves as a foundation for a new relation: emph{likely-happened-before} denoted by $xrightarrow{p}$ where $p$ represents the probability of an event to have happened before another. The $xrightarrow{p}$ relation provides a basis for ordering multiple events which are otherwise considered emph{concurrent} by the typical emph{happened-before} ($ ightarrow$) relation. We highlight various related challenges including intransitivity of $xrightarrow{p}$ relation as opposed to the transitive $ ightarrow$ relation. We also outline several research directions: online fair sequencing, stochastically fair total ordering, host-level support for fairness and more.

Problem

• Achieving fair event ordering despite clock synchronization limitations
• Using statistical models to compare noisy timestamps probabilistically
• Establishing likely-happened-before relations for ordering concurrent events

Innovation

• Leverages statistical model for timestamp comparison
• Computes event precedence probability without wall-clock
• Introduces likely-happened-before relation for ordering