III: Small: On the Conceptual Evaluation and Optimization of Queries in Spatiotemporal Data Systems

这个项目的目标是研究和开发正确的概念评估、处理和优化多谓词空间(MPS)查询的技术，以支持基于位置的服务。该项目通过以下方法实现其目标：(1)研究MPS查询的特点，为MPS查询开发一种新的一致且可证明正确的概念评估策略，类似于关系SQL查询；(2)开发各种查询转换规则，不仅保持正确，而且将MPS查询执行计划转换为更高效的查询执行计划；(3)研究著名的关系优化启发式算法在MPS查询环境中的有效性，并开发针对MPS查询的创新优化算法；(4)开发了一个代价模型来帮助查询优化器为MPS查询选择最优的查询执行计划，并研究了时空移动对象分布的变化对MPS查询执行计划的影响；(5)开发了能够应对移动对象环境的动态性质和规模的新的自适应时空查询处理技术；(6)原型了一个反映连续和自适应MPS算子的正确评估和高效优化执行的MPS位置服务器；(7)开发了模拟器和分析模型来评估所开发技术的性能。研究成果对位置服务器、智能交通系统、智慧城市、供应链管理、应急响应与恢复等领域的应用具有重要意义。该项目支持博士生在时空数据管理系统和高级位置服务器领域进行研究，并邀请本科生参与相关研究项目。该项目开发和引入了一门关于2D和3D地图操作和查询的新的入门级本科编程课程，该课程利用MPS查询并整合了该项目的研究成果，以提高计算机科学和其他科学学科的学生对编程的兴趣。该项目通过普渡探索公园的发现学习中心、路易斯·斯托克斯少数族裔参与联盟(LSAMP)项目和研究生教育计划联盟(AGEP)让少数族裔学生参与。这项研究的出版物、技术报告、软件和实验数据可在项目网站(http://www.cs.purdue.edu/~aref/MPS).)上获得

The goal of this project is to research and develop techniques for the correct conceptual evaluation, processing, and optimization of multi-predicate spatial (MPS) queries in support of location-based services. The project achieves its goal using the following approaches: (1) Study the characteristics of MPS queries and develop a new consistent and provably correct conceptual evaluation strategy for MPS queries similar to the one for relational SQL queries; (2) Develop various query transformation rules that not only retain correctness but also transform MPS query execution plans into more efficient ones; (3) Investigate the validity of well-known relational optimization heuristics in the context of MPS queries and develop innovative optimization algorithms unique to MPS queries; (4) Develop a cost model to aid the query optimizer in selecting the best query execution plans for MPS queries and study the impact of the variation on temporal and spatial moving object distributions on MPS query execution plans; (5) Develop new adaptive spatiotemporal query processing techniques that can cope with the dynamic nature and scale of moving object environments; (6) Prototype an MPS location server that reflects the correct evaluation and efficient optimized execution of continuous and adaptive MPS operators; and (7) Develop simulators and analytical models to evaluate the performance of the developed techniques. The research results are beneficial to many applications including location servers, intelligent transportation systems, smart cities, supply chain management, and emergency response and recovery.This project supports Ph.D. students to pursue research in the areas of spatiotemporal data management systems and advanced location servers, and involves undergraduate students in related research projects. The project develops and introduces a new entry-level undergraduate programming course around 2D and 3D map operations and queries that utilizes MPS queries and integrates the research results from the project to increase the interest in programming of students from Computer Science as well as from other Science disciplines. The project involves minority students through the Discovery Learning Center at Purdue Discovery Park and the Louis Stokes Alliance for Minority Participation (LSAMP) project and Alliances for Graduate Education Program (AGEP). Publications, technical reports, software and experimental data from this research are available at the project web site (http://www.cs.purdue.edu/~aref/MPS).