arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1601.04749cs.NIcs.PF
收藏

约束多用户多服务器Max-Min公平调度

Constrained Multi-user Multi-server Max-Min Fair Queuing

Jalal Khamse-Ashari, Ioannis Lambadaris, Yiqiang Zhao

摘要:

本文研究了一个多用户多服务台排队系统,其中每个用户被限制只能从一个服务台子集获得服务。在所研究的系统中,在最大-最小公平性意义上的速率分配导致多级公平速率。为了实现这样的公平速率,我们提出了$CM^4FQ$算法。在该算法中,在逐包的基础上为服务选择用户。在时间$t$选择的每个用户$i$的优先级基于称为服务标签的参数(表示到时间$t$为止对用户$i$计数的工作量)来确定。因此,免费服务器将选择为具有最小服务标签的合格用户提供服务。基于这种简单的选择标准,$CM^4FQ$的目标是保证公平的吞吐量为每个需求的用户没有明确的知识,每个服务器的服务速率。我们认为,$CM^4FQ$可以应用于各种实际的排队系统,特别是在移动云计算架构。

英文摘要:

In this paper, a multi-user multi-server queuing system is studied in which each user is constrained to get service from a subset of servers. In the studied system, rate allocation in the sense of max-min fairness results in multi-level fair rates. To achieve such fair rates, we propose $CM^4FQ$ algorithm. In this algorithm users are chosen for service on a packet by packet basis. The priority of each user $i$ to be chosen at time $t$ is determined based on a parameter known as service tag (representing the amount of work counted for user $i$ till time $t$). Hence, a free server will choose to serve an eligible user with the minimum service tag. Based on such simple selection criterion, $CM^4FQ$ aims at guaranteed fair throughput for each demanding user without explicit knowledge of each server service rate. We argue that $CM^4FQ$ can be applied in a variety of practical queuing systems specially in mobile cloud computing architecture.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑