CRII: CNS: With Uncertainty Comes Opportunity: Providing Best-Effort Edge Services with Uncertain and Limited Resources

这个项目的目标是创建一个运行时系统，以提供最佳的边缘服务。与具有可管理和冗余资源的云不同，许多边缘环境是机会性的，具有不确定和有限的资源。因此，边缘服务无法保证其客户端的服务质量（QoS），这是创建当前和新兴边缘计算应用程序的关键障碍。现有的解决方案存在不足，因为它们通过分配服务竞争的瓶颈资源来优化总体QoS。相比之下，本项目基于边缘环境的不确定性也带来了机会，因此不消耗相同资源的等效功能可以满足相同的服务请求。因此，本项目将探索如何通过动态编排消耗未充分利用的资源的等效功能来系统地改进QoS。该项目开发的技术将加速边缘计算在自动驾驶汽车、智能家居和工业4.0等新兴领域的大规模应用。一个特定的目标是使在边缘计算环境中使用任何类型的硬件设备成为可能。为这个项目开发的软件基础设施将用于教学目的，创建新的和丰富现有的分布式系统和边缘计算课程。该项目在底特律大都会区执行，为扩展到当地不同社区提供了绝佳的机会，从而加速了终端用户对边缘计算技术的采用，并吸引了代表性不足的群体参与计算学习和研究。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The goal of this project is to create a runtime system for providing best-effort edge services. Different from clouds with manageable and redundant resources, many edge environments are opportunistic and feature uncertain and limited resources. As a result, edge services cannot guarantee Quality of Service (QoS) for their clients, which is a critical obstacle of creating current and emerging edge computing applications. Existing solutions fall short, as they optimize the overall QoS by allocating bottleneck resources competed for by services. In contrast, this project is based on the insight that the uncertainty of edge environments also brings opportunity so the equivalent functions that do not consume the same resources can satisfy the same service request. Hence, this project will explore how to systematically improve QoS by dynamically orchestrating equivalent functions that consume underutilized resources.The technologies developed during the course of this project will accelerate large-scale applications of edge computing in emerging domains like autonomous vehicles, smart homes, and Industry 4.0. A particular objective is to make it possible to use any kinds of hardware devices in edge computing environments. The software infrastructure developed for this project will be used for instructional purposes to create new and enrich existing courses on distributed systems and edge computing. Executed in the Detroit Metropolitan area, this project provides an excellent opportunity for outreach to local diverse communities, thus accelerating end-user adoption of edge computing technologies and attracting underrepresented groups into computing study and research.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.