CSR: Small: Data on the Edge: Leveraging Edge Datacenters for Low-latency, Fault-tolerant, mobile Geo-replicated Transactional Data Stores

CSR：小型：边缘数据：利用边缘数据中心实现低延迟、容错、移动地理复制事务数据存储

• 批准号: 1815733
• 负责人: Divyakant Agrawal
• 金额: $ 24.99万
• 依托单位: University of California-Santa Barbara
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-10-01 至 2021-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1815733&HistoricalAwards=false
• 关键词: CSR; Small; Data; Edge; Leveraging

Data-intensive computing is the driver of the continuing growth and success of global Internet services and cloud applications. The system infrastructure for data-intensive computing is built and run on top of compute resources housed in large, centralized data centers. These data centers are typically placed away from populous regions for economic reasons. This leads to a fundamental limit on the speed of communication between users and data centers. To overcome this limit, the project aims to extend data management systems to be edge-aware. Edge-awareness is the capability of utilizing resources that are closer to users to reduce the latency they experience by one to two orders of magnitude.The project will investigate two closely related research challenges to enable edge-aware data management systems. The first research challenge focuses on one of the most important building blocks of distributed systems, namely the problem of agreement between different nodes (also called the consensus problem.) The project proposes a design of a new consensus protocol that, unlike existing consensus protocols, is edge-aware and thus enables preserving consensus and fault-tolerance while achieving the low-latency performance enabled by edge-awareness. The second research challenge is building edge data management systems that react efficiently to user mobility. The project proposes augmenting consensus protocols with a mobility component that enables handling user mobility efficiently while preserving consensus and fault-tolerance.Large sectors of business and industry, in addition to the modern style of living, increasingly rely on data-intensive Internet and cloud applications. This project aims to advance the data and system infrastructure for these applications. The project will also provide the opportunity for undergraduate and graduate students to learn and work on fundamental data management systems 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.

数据密集型计算是全球互联网服务和云应用持续增长和成功的驱动力。数据密集型计算的系统基础架构是在大型集中式数据中心的计算资源上构建和运行的。出于经济原因，这些数据中心通常远离人口稠密的地区。这导致用户和数据中心之间的通信速度受到根本限制。 为了克服这一限制，该项目旨在将数据管理系统扩展到边缘感知。 边缘感知是利用更接近用户的资源，将用户体验的延迟减少一到两个数量级的能力。该项目将研究两个密切相关的研究挑战，以实现边缘感知数据管理系统。第一个研究挑战集中在分布式系统最重要的构建模块之一，即不同节点之间的一致性问题（也称为共识问题）。该项目提出了一种新的共识协议的设计，与现有的共识协议不同，该协议是边缘感知的，因此能够保持共识和容错，同时实现边缘感知所实现的低延迟性能。第二个研究挑战是构建边缘数据管理系统，对用户移动性做出有效反应。该项目建议在共识协议中增加移动组件，以便在保持共识和容错的同时有效地处理用户移动性。除了现代生活方式之外，大规模的商业和工业部门越来越依赖于数据密集型互联网和云应用。 该项目旨在推进这些应用程序的数据和系统基础设施。该项目还将为本科生和研究生提供学习和从事基础数据管理系统研究的机会。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。

项目成果

Permissioned Blockchains: Properties, Techniques and Applications

• DOI: 10.1145/3448016.3457539
• 发表时间: 2021-06
• 期刊: Proceedings of the 2021 International Conference on Management of Data
• 作者: M. Amiri;D. Agrawal;A. E. Abbadi

SharPer: Sharding Permissioned Blockchains Over Network Clusters

• DOI: 10.1145/3448016.3452807
• 发表时间: 2019-10
• 期刊: Proceedings of the 2021 International Conference on Management of Data
• 作者: Mohammad Javad Amiri;D. Agrawal;A. E. Abbadi

Coeus: A System for Oblivious Document Ranking and Retrieval

• DOI: 10.1145/3477132.3483586
• 发表时间: 2021-10
• 期刊: Proceedings of the ACM SIGOPS 28th Symposium on Operating Systems Principles
• 作者: Ishtiyaque Ahmad;Laboni Sarker;D. Agrawal;A. E. Abbadi;Trinabh Gupta

Fides: Managing Data on Untrusted Infrastructure

Fides：管理不可信基础设施上的数据

• DOI: 10.1109/icdcs47774.2020.00053
• 发表时间: 2020
• 期刊: IEEE International Conference on Distributed Computing Systems
• 作者: Maiyya, Sujaya;Bum Cho, Danny Hyun;Agrawal, Divyakant;El Abbadi, Amr
• 通讯作者: El Abbadi, Amr

Unifying Consensus and Atomic Commitment for Effective Cloud Data Management

• DOI: 10.14778/3303753.3303765
• 发表时间: 2019-01
• 期刊: Proc. VLDB Endow.
• 作者: Sujaya Maiyya;Faisal Nawab;D. Agrawal;A. E. Abbadi