Research Infrastructure: CC* Data Storage: Rice Collaborative Object Store

研究基础设施：CC* 数据存储：Rice 协作对象存储

• 批准号: 2322372
• 负责人: Erik Engquist
• 金额: $ 20.6万
• 依托单位: William Marsh Rice University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-08-01 至 2025-07-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2322372&HistoricalAwards=false
• 关键词: Research; Infrastructure; CC; Data; Storage

Research in areas such as modeling and simulation, AI and machine learning, large-scale geospatial analysis, automated data mining of large digital text and image collections, and bioengineering produce terabyte and petabyte-scale data sets that require production-level storage, access, and transfer services. The Rice Collaborative Object Store adds an S3 compatible resource to on-campus cyberinfrastructure services, providing much needed capacity for multi-terabyte data, making that data available for local -or distant- computational analysis, sharing with distributed collaborations, or available for public access. As the research computing and data applications expand across fields, there is a need to create opportunities for workforce development in the area of cyberinfrastructure. This project introduces students to the advanced computing ecosystem, providing training on local storage and compute resources, and user facilitation. These student trainees actively participate in research engagement with the collaborating use case groups to utilize this new resource and the Open Storage Network (OSN).The Rice Collaborative Object Store node joins the production-level OSN, which is part of the nation’s growing federated data fabric. Providing low-cost, efficient services based on open-source software, the OSN has nodes at many of the national or regional advanced computing centers. The system consists of three compute nodes in a high availability configuration, 1.9PB disk, and Ceph storage software, connected at 100 gigabits from the Rice University ScienceDMZ network. The resource contributes 20% of the available storage to the common OSN allocation pool that is available to researchers across the U.S.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.

建模和仿真、人工智能和机器学习、大规模地理空间分析、大型数字文本和图像集的自动数据挖掘以及生物工程等领域的研究产生了TB和PB级的数据集，这些数据集需要生产级的存储、访问和传输服务。Rice Collaborative Object Store为校园网络基础设施服务增加了S3兼容资源，为多TB数据提供了急需的容量，使这些数据可用于本地或远程计算分析，与分布式协作共享，或可供公众访问。随着研究计算和数据应用在各个领域的扩展，有必要为网络基础设施领域的劳动力发展创造机会。该项目向学生介绍先进的计算生态系统，提供本地存储和计算资源以及用户便利方面的培训。这些学生学员积极参与协作用例组的研究工作，以利用这一新资源和开放存储网络（OSN）。Rice协作对象存储节点加入了生产级OSN，后者是美国不断发展的联邦数据结构的一部分。OSN基于开源软件提供低成本、高效率的服务，在许多国家或地区的先进计算中心都有节点。该系统由三个高可用性配置的计算节点、1.9PB磁盘和Ceph存储软件组成，以100千兆位从Rice University ScienceDMZ网络连接。该资源将20%的可用存储贡献给美国各地研究人员可用的公共OSN分配池。该奖项反映了NSF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。