PPoSS: Planning: Dynamic Big Graph Store for High-Throughput and Secure Distributed Query Processing

PPoSS：规划：用于高吞吐量和安全分布式查询处理的动态大图存储

• 批准号: 2028714
• 负责人: Rajiv Gupta
• 金额: $ 25万
• 依托单位: University of California-Riverside
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2021-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2028714&HistoricalAwards=false
• 关键词: PPoSS; Planning; Dynamic; Big; Graph

Since graphs can readily express entities and relationships among them, they are widely used to represent large volumes of data from domains ranging from online shopping platforms to social networks. These graphs continue to evolve in size and structure as they integrate new data that emerges, often in real time, thus creating a need for online analytics to answer user queries based upon the latest available information. The goal of this research is to address fundamental challenges across the software-hardware spectrum to develop a scalable and secure distributed platform for online graph analytics. This research will discover novel techniques for delivering meaningful answers to queries over changing data, with a high degree of efficiency, while addressing security concerns that can arise in cloud settings. The key components of this platform, named DyGr, include: a graph store with a consistency model that tailors the scope of consistency to efficiently compute meaningful answers to queries; an event-driven incremental computation model that is supported across software and hardware layers to rapidly compute query results; and hardware supported protocols for secure distributed evaluation of queries over private data. By building of a powerful online analytics system this research will contribute to acceleration of new discoveries in fields that employ graph analytics. Advances in online analytics will enable new applications that are relevant to businesses and thus will contribute to economic growth. The students participating in this project will gain comprehensive training in all aspects of system building as well as receive training in applications that require online analytics. Therefore, the broader impact in form of contributions to workforce development in an area of national need will be tremendous.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.

由于图形可以很容易地表达实体及其之间的关系，因此它们被广泛用来代表从在线购物平台到社交网络的域中的大量数据。这些图的大小和结构不断发展，因为它们整合了经常实时出现的新数据，因此需要根据最新的可用信息来回答用户查询。这项研究的目的是解决软件硬件频谱中的基本挑战，以开发可扩展且安全的分布式平台，以用于在线图分析。这项研究将发现新颖的技术，可以通过高度效率为不断变化的数据提供有意义的查询答案，同时解决在云设置中可能出现的安全问题。该平台的关键组件（名为Dygr）包括：具有一致性模型的图表存储，可定制一致性的范围，以有效地计算出查询的有意义的答案；事件驱动的增量计算模型，在软件和硬件层之间得到支持，以快速计算查询结果；和硬件支持的协议，用于通过私人数据对查询的安全分布式评估。通过建立强大的在线分析系统，这项研究将有助于加速采用图形分析的领域。在线分析的进步将实现与企业相关的新应用程序，从而有助于经济增长。参加该项目的学生将在系统构建的各个方面获得全面的培训，并在需要在线分析的应用程序中接受培训。因此，在国家需求领域对劳动力发展的贡献形式的更广泛影响将是巨大的。该奖项反映了NSF的法定任务，并被认为是值得通过基金会的知识分子优点和更广泛的审查标准通过评估来获得支持的。

项目成果

SimGQ: Simultaneously Evaluating Iterative Graph Queries

• DOI: 10.1109/hipc50609.2020.00014
• 发表时间: 2020-12
• 期刊: 2020 IEEE 27th International Conference on High Performance Computing, Data, and Analytics (HiPC)
• 作者: Chengshuo Xu;Abbas Mazloumi;Xiaolin Jiang;Rajiv Gupta

VRGQ: Evaluating a Stream of Iterative Graph Queries via Value Reuse

VRGQ：通过值重用评估迭代图查询流

• DOI: 10.1145/3469379.3469382
• 发表时间: 2021
• 期刊: ACM SIGOPS Operating Systems Review
• 作者: Jiang, Xiaolin;Xu, Chengshuo;Gupta, Rajiv
• 通讯作者: Gupta, Rajiv

BEAD: Batched Evaluation of Iterative Graph Queries with Evolving Analytics Demands

• DOI: 10.1109/bigdata50022.2020.9378211
• 发表时间: 2020-12
• 期刊: 2020 IEEE International Conference on Big Data (Big Data)
• 作者: Abbas Mazloumi;Chengshuo Xu;Zhijia Zhao;Rajiv Gupta

Tripoline: generalized incremental graph processing via graph triangle inequality

• DOI: 10.1145/3447786.3456226
• 发表时间: 2021-04
• 期刊: Proceedings of the Sixteenth European Conference on Computer Systems
• 作者: Xiaolin Jiang;Chengshuo Xu;Xizhe Yin;Zhijia Zhao;Rajiv Gupta

GraphPulse: An Event-Driven Hardware Accelerator for Asynchronous Graph Processing

• DOI: 10.1109/micro50266.2020.00078
• 发表时间: 2020-10
• 期刊: 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)
• 作者: Shafiur Rahman;N. Abu-Ghazaleh;Rajiv Gupta