Approximate structures for efficient processing of data streams

高效处理数据流的近似结构

• 批准号: DP190102078
• 负责人: Prof Justin Zobel
• 金额: $ 27.76万
• 依托单位: The University of Melbourne
• 依托单位国家: 澳大利亚
• 项目类别: Discovery Projects
• 财政年份: 2019
• 资助国家: 澳大利亚
• 起止时间: 2019-01-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://dataportal.arc.gov.au/RGS/Web/Grants/DP190102078
• 关键词: Approximate; structures; efficient; processing; data

This project aims to increase the volume of streamed data that can be handled on a low-powered device with limited memory. In finance, health, and transport, data arrives at enormous rates, and data-driven decisions must be made quickly. Likewise, to keep Australia secure, national agencies monitor and gather vast data sets. Increasingly, devices and monitors that have limited resources are making these decisions and they require computational techniques that run extremely efficiently. The project expects to develop and improve approximate data structures that operate in tight resource bounds. Anticipated outcomes are improved event recognition and dramatic speedup in analysis of streams in areas such as finance, health, transport, and urban data.

该项目旨在增加可以在内存有限的低功耗设备上处理的流数据量。在金融、卫生和交通领域，数据的到达速度非常快，必须快速做出数据驱动的决策。同样，为了确保澳大利亚的安全，国家机构监控和收集大量数据集。越来越多的设备和监视器资源有限，他们需要非常高效的计算技术来做出这些决定。该项目预计将开发和改进在严格的资源范围内运行的近似数据结构。预期的结果是改进事件识别，并大幅加快金融、卫生、交通和城市数据等领域的流分析。