EAGER: Designing the OSN Software Platform

该奖项将探索为一种全新类型的基础设施提供动力所需的软件平台，即国家规模的开放存储网络。该项目团队将与位于约翰霍普金斯大学的开放存储网络（OSN）项目团队和四个NSF大数据区域创新中心密切合作，后者将收购和运营OSN的硬件组件。该项目将涉及计算机科学家、软件和网络工程师以及应用科学家之间的多学科合作，以开发和原型软件。目标是快速了解软件平台上的关键要求，为OSN的下一阶段做准备，这将涉及部署一个全国性的、生产质量的存储网络，该网络具有数百pb的容错、分布式存储，通过高速网络相互连接。本项目重点设计和开发OSN软件平台和相应的编程接口，确保三种用户群体能够高效、便捷地访问：OSN系统管理员，负责监控和管理OSN的运行；应用程序开发人员，他们开发基于OSN功能的创新应用程序；以及直接或通过此类应用程序使用OSN的研究人员、教育工作者、学生和其他人员。该资助探索和开发开放存储网络（OSN）节点软件堆栈的快速原型，以回答对国家规模实施的最终成功至关重要的探索性问题，例如：-应该为OSN这样的系统的设计提供哪些用例？-管理OSN这样的系统需要什么样的机制？-在核心OSN结构的基础上，需要提供哪些增值服务来支持应用用例？-可持续性模型可用于扩展和维持OSN？- OSN用例和工作流程对存储系统级OSN基础设施有什么影响？为了为这种探索性的、快速的原型工作提供基础，EAGER团队将与OSN团队合作，建立一个功能存储测试平台，该平台具有足够强大的软件，可以给用户信心，并能够在真正的科学应用中使用，这将是该测试平台的测试用例。这项工作将包括部署现有的软件，如Ceph和Globus。然而，这项工作只是为解决上述探索性问题做准备，这些问题是拟议项目的重点。这里提出的工作将原型化OSN软件平台的初始版本，该平台提供了一套初始的身份验证、授权、数据移动和数据共享功能，以支持OSN实验和验证。该项目还将根据社区意见，并以早期实验研究为基础，为其下一个版本设计和开发一个功能更全面的OSN软件平台。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。

This EAGER award will explore the software platform needed to power a radically new type of infrastructure, viz., a national-scale open storage network. The project team will work closely with the Open Storage Network (OSN) project team, located at Johns Hopkins University and the four NSF Big Data Regional Innovation Hubs, who will acquire and operate the hardware component of the OSN. The project will engage in multidisciplinary collaboration among computer scientists, software and network engineers, and application scientists, to develop and prototype the software. The goal is to learn rapidly about the key requirements placed on the software platform in preparation for the next phase of OSN, which would involve the deployment of a nationwide, production-quality storage network with hundreds of petabytes of fault tolerant, distributed storage, interconnected via high-speed networks. The focus of this project is on the design and development of the OSN software platform and corresponding programming interfaces, to ensure that they can be accessed effectively and easily by three groups of users: OSN system administrators, who are required to monitor and manage OSN operations; application developers, who develop innovative applications that build upon OSN capabilities; and researchers, educators, students, and others, who use the OSN either directly or through such applications. This grant explores and develops rapid prototypes of the software stack for nodes of the Open Storage Network (OSN), with a view to answering exploratory questions vital to the eventual success of a national-scale implementation, such as:-What are the use cases that should inform the design of a system like OSN?-What are the mechanisms that will be required to administer and manage a system like OSN?-What value-added services should be provided on top of a core OSN fabric to support application use cases?-What sustainability models can be used to expand and sustain OSN?-What are the implications of OSN use cases and workflows for OSN infrastructure at the storage system level?To provide a foundation for this exploratory, rapid prototyping effort the EAGER team will work with the OSN team to establish a functional storage testbed with software that is sufficiently robust to give users confidence and enable use in real scientific applications, which will be the test cases for this testbed. That work will involve deployment of existing software such as Ceph and Globus. However, that work is just preparatory to addressing the exploratory questions listed above, which are the focus of the proposed project. The work proposed here will prototype an initial version of the OSN Software Platform that provides an initial set of authentication, authorization, data movement, and data sharing capabilities to support OSN experimentation and validation. The project will also work towards the design and development of a more full-featured OSN Software Platform for its next version, based upon community input, and backed up by early experimental studies.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.