SMA/PDOS Collaborative Research: Design, Analysis, and Control of Adaptive Sharing Mechanisms

各种各样的系统，包括网络场、虚拟机、多任务操作系统、网格计算系统和传感器网络，通过在它们支持的消费者之间“共享”资源，提高了它们的可访问性、可用性、弹性和公平性。然而，探索如何共享资源的研究通常会得出要点解决方案，其中不同的资源/消费者配置需要单独设计的共享机制。例如，调度器通常单独实现单个策略(例如，FCFS、PS、前台后台、最短剩余处理器时间)，针对特定的负载设置进行优化，并且当情况改变时不能容易地切换到另一个策略。该项目正在开发和分析自适应共享机制(ASM)，其中用于共享资源的机制动态地适应可用资源集和消费者的当前需求，从而使得系统是真正自治的。这项研究是以将军事管理系统模块化为独立的组成部分开始的，这些组件由运作的时间尺度确定。利用前沿的控制理论和调度分析对各个组件进行了研究。ASM的设计和分析将为完全自主的系统提供理论基础，该项目将导致不同系统中资源的有效利用，如服务器群、网格计算和传感器网络。

A wide variety of systems, including web farms, virtual machines,multi-tasking OSes, GRID computing systems, and sensor networksimprove their accessibility, availability, resilience and fairness by``sharing'' resources across the consumers they support. However,research that explores how to share resources generally derives pointsolutions, where different resource/consumer configurations requireseparately-designed sharing mechanisms. For instance, a scheduleroften has implemented separately a single policy(e.g., FCFS, PS, Foreground Background, Shortest Remaining ProcessorTime), optimized for a particular load settingand cannot easily be switched to another policy when the situation changes.This project is developing and analyzing Adaptive SharingMechanisms (ASMs) in which the mechanism used to share resourcesadapts dynamically to both the set of available resources and thecurrent needs of the consumers, such that the system is trulyautonomic. The study has been initiated with a modularization of the ASMinto separate components, defined by the timescale of operation. Astudy of the various components using both cutting edge novel controltheoretic and scheduling analyses is being performed.The design and analysis of ASMs will provide a theoreticalgrounding for fully autonomic systems, and the project will lead toefficient utilization of resources in diverse systems like serverfarms, GRID computing and Sensor Networks.