ITR: Virtual Power for a Wireless Campus - A Vision of Ubiquitous Computing On Low-Cost Mobile Devices

虽然手持、电池供电的设备，如个人数字助理(PDA)和支持网络的手机正在崛起，作为新的接入指向世界的数字基础设施，但它们的成本和较短的电池寿命是阻碍其巨大潜力的因素。更糟糕的是，这类设备的成本甚至可能扩大“数字鸿沟”。这项研究同时解决了这些成本和电池寿命问题，从而离普适计算涵盖整个社会的愿景又近了一步。具体的焦点是具有无线互联网覆盖的数字大学校园。在这种背景下，目的是通过使移动设备能够使用相对简单的硬件来增加手持无线计算机的实用性和电池寿命，并降低成本。这项研究旨在设计能够更好地节约资源的嵌入式硬件，以及创建一个软件层，通过无线连接将手持设备无缝地耦合到功率相对较高的固定计算基础设施，从而掩盖手持设备有限的计算能力。通过将耗电密集的操作转移到固定基础设施，电池供电的移动设备将获得“虚拟电力”。将开发自适应即时编译器技术，以最大限度地减少运行移动代码的移动设备的能耗，并利用计算网格研究社区的结果开发自适应调度方法。根据要在移动设备上运行的算法及其与最近基站的当前距离，要执行的计算将被自动地划分为在固定基础设施中执行的部分和将在移动设备上运行的部分。在硬件级别，将制定协调的资源管理策略，以使设计人员能够正确地设计和实现高度资源受限的嵌入式系统，同时帮助他们满足系统级别的限制。这需要用以资源为中心的观点来增强当今的功能设计流程。这里的目标不是用另一种无所不包的方法来取代现有的设计方法，而是通过展示结果对不同抽象级别的几个驱动实例的适用性来产生交叉影响，包括片上系统(SoC)平台、内存体系结构级别和操作系统级别。

While handheld, battery-powered devices such as personal digitalassistants (PDA's) and web-enabled mobile phones are emerging as newaccess points to the world's digital infrastructure, their cost andshort battery life are factors that are holding back their enormouspotential. Worse yet, the cost of such devices might even widen the"digital divide". This research addresses these cost and battery-life issuessimultaneously, thereby getting one step closer to a vision of ubiquitouscomputing embracing all of society. The specific focus is the digital universitycampus with wireless Internet coverage.In this setting, the aim is to increase the utility and battery-lifeand decrease the cost of handheld wireless computers by enabling theuse of relatively simple hardware for the mobile devices. This research aimsboth at designing embedded hardware that better conserves resources, as well ascreating a software layer that masks the limited computational prowessof a handheld device by seamlessly coupling it to a relatively high-poweredstationary computational infrastructure via an "always on" wireless connection.By off-loading power-intensive operations to the stationary infrastructure,the battery-powered mobile device is provided with "virtual power".Adaptive just-in-time compiler technology will be developed for minimizingpower use on mobile devices running mobilecode, and adaptive scheduling methods using results from theComputational Grid research community. Depending on the algorithm tobe run on the mobile device and its current distance from the nearestbase station, the computation to be performed is to be automaticallypartitioned between a part to be executed in the stationaryinfrastructure and another to be run on the mobile device.At the hardware level, orchestrated resource-management strategies will bedeveloped to enable designers to correctly design and implement highlyresource-constrained embedded systems while helping them to meet system-levelconstraints. This requires augmenting today's functional design flows with aresource-centric view. Here, the goal is not to replace existing design methodologieswith yet another all-encompassing methodology, but rather making a cross-cuttingimpact by demonstrating the applicability of results to several driving examplesat different levels of abstraction, including System-on-Chip (SoC) platforms,memory architecture level, and operating system level.