[论文解读] A Quantitative Measure Of Fairness And Discrimination For Resource Allocation In Shared Computer Systems

Rishabh Jain, Dah Ming Chiu+1 more|ArXiv.org|Sep 24, 1998

Cloud Computing and Resource Management被引用 3,564

一句话总结

本文提出一种通用、与资源量无关的公平性度量，称为 Indiex of FRairness，其歧视指数定义为 1 减去 fairness index，适用于任何资源共享问题。

ABSTRACT

Fairness is an important performance criterion in all resource allocation schemes, including those in distributed computer systems. However, it is often specified only qualitatively. The quantitative measures proposed in the literature are either too specific to a particular application, or suffer from some undesirable characteristics. In this paper, we have introduced a quantitative measure called Indiex of FRairness. The index is applicable to any resource sharing or allocation problem. It is independent of the amount of the resource. The fairness index always lies between 0 and 1. This boundedness aids intuitive understanding of the fairness index. For example, a distribution algorithm with a fairness of 0.10 means that it is unfair to 90% of the users. Also, the discrimination index can be defined as 1 - fairness index.

研究动机与目标

• 将公平性作为资源分配中的关键性能指标予以强调。
• 批评现有的定性或应用场景特定的公平性度量。
• 引入一个通用、与资源无关的公平性指数，该指数是有界且可解释的。

提出的方法

• 将 Indiex of FRairness 定义为通用的公平性度量。
• 确保 fairness index 与资源量无关。
• 使 fairness index 的取值在 0 与 1 之间，以便直观解释。
• 解释：fairness of 0.10 表示对 90% 用户的不公平。
• 将 discrimination index 定义为 1 减去 fairness index。

实验结果

研究问题

• RQ1 是否能够为任何资源共享问题定义一个通用、资源无关的公平度量？
• RQ2 在不同分布下，公平性应如何有界并被解释？
• RQ3 资源分配中的公平性与歧视之间的关系是什么？

主要发现

• 提出一个在 0 与 1 之间有界的公平性指数。
• 将 discrimination index 定义为 1 减去 fairness index。
• 提供对公平性数值的直观解释（例如，数值越低表示不公平程度越高）。