SHF: Small: Resilient Operations for Deployment Units Used in Container Orchestration

在软件工程中，容器是一个自包含的软件包，它包含在任何环境中运行的所有必要元素。为了向客户快速部署软件，组织使用容器编排。容器编排是一种实际管理大规模容器的实践，其中可以自动供应和管理数百个容器，即轻量级计算机虚拟化工具。到2026年，容器编排的市场价值预计将达到13亿美元。虽然这种实践已经产生了好处，但是容器编排的弹性仍然是组织面临的一个挑战。如果没有容器编排的弹性实现，部署的软件就会变得不可靠，从而给组织及其客户带来大规模的后果。由于容器编排有其独特的特点，有必要进行基础的科学研究。这些研究调查的执行将产生新的科学证据，产生容器编排特有的新技术，并推进软件工程研究的状态。该项目生成的所有软件都将公开提供给行业从业者、研究软件工程师和学者使用。项目产生的数据集将有利于整个软件工程界进行进一步的研究。该项目还将通过积极参与弱势群体的学生，促进STEM的广泛参与。通过将生成的研究成果传播到研究生和本科课程，学生将获得有关弹性容器编排的知识，这将直接有助于美国国家战略计算计划所强调的发展知识渊博的网络基础设施劳动力的国家持续努力。该项目将开展研究活动，调查如何将弹性集成到容器编排的实践中。这个项目的重点是发展部署单元的弹性操作科学，这是执行容器编排的关键。该项目将执行两项任务：(i)通过检测差异状态和弹性卷管理来弹性地提供部署单元，以及（ii）通过检测防火墙和区域差异来弹性地路由部署单元之间的流量。该项目将确定部署、数量和区域差异的根本原因，这些差异是容器编排所特有的。本项目由CISE软硬件基金项目和EPSCoR （Established program to incentive Competitive Research）共同资助。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。

In Software Engineering, a Container is a self-contained package of software that includes all of the necessary elements to run in any environment. To deploy software rapidly to customers, organizations use container orchestration. Container orchestration is the practice of pragmatically managing containers at scale, where hundreds of containers, i.e., lightweight computer virtualization tools, can be automatically provisioned and managed. Container orchestration is expected to reach USD 1.3 billion worth of market value by 2026. While this practice has yielded benefits, resiliency for container orchestration remains a challenge for organizations. Without resilient implementation of container orchestration, deployed software becomes unreliable, creating large-scale consequences for organizations and their customers. As certain characteristics are unique to container orchestration, foundational scientific research investigations are necessary. Execution of these research investigations will yield new scientific evidence, generate novel techniques unique to container orchestration, and advance the state of software engineering research. All software generated from the project will be publicly available to industry practitioners, research software engineers, and academics for usage. Datasets generated from the project will be beneficial to the entire software engineering community to conduct further research. The project will also foster broadening of participation in STEM by pro-actively involving students from under-represented groups. Through dissemination of generated research into graduate and undergraduate courses, students will gain knowledge on resilient container orchestration, which will directly contribute to the nation's ongoing efforts in developing a knowledgeable cyber infrastructure workforce, emphasized by the U.S. National Strategic Computing Initiative.This project will conduct research activities to investigate how to integrate resiliency into the practice of container orchestration. The focus of this project is to grow the science of resilient operations for deployment units, which are pivotal to perform container orchestration. This project will execute two tasks: (i) resilient provisioning of deployment units through detection of discrepant states and resilient volume management, and (ii) resilient routing of traffic between deployment units through detection of firewall and zonal discrepancies. This project will identify root causes of deployment, volume, and zonal discrepancies that are unique to container orchestration.This project is jointly funded by CISE Software and Hardware Foundations program and the Established Program to Stimulate Competitive Research (EPSCoR).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.