Scaling Disk-Resident Learned Indexes For Database Systems

这个项目旨在研究新的驻留在磁盘上的学习索引算法，以通过推进驻留在内存中的学习建模的最新技术来存储和处理数据库系统中的数据。该项目期望在数字存储技术领域产生新的知识，利用新的、高效的技术对大数据进行学习索引。这应该会带来显著的好处，使现代数据库系统能够随着数据的大规模增长而扩展，提高数据处理的效率，提高利用大数据的项目的效率，并显著降低澳大利亚数据中心在从数据库存储和检索数据时的能源成本，并降低它们的碳足迹。

This project aims to investigate new disk-resident learned indexing algorithms to store and process data in database systems by advancing the state-of-the-art in memory-resident learned modeling. This project expects to generate new knowledge in the area of digital storage technologies utilising novel and efficient techniques in learned indexing for big data. This should provide significant benefits to enable modern database systems to scale with the massive growth of data, improve the efficiency of data processing, improve the effectiveness of projects that utilise big data, and dramatically reduce energy costs in Australian data centres when storing and retrieving data from databases and lower their carbon footprints.