IIIT Hyderabad Publications |
|||||||||
|
Performance Analysis of Spatial Queries and Routing in NoSQL databasesAuthor: Sarthak Agarwal Date: 2019-04-09 Report no: IIIT/TH/2019/37 Advisor:K S Rajan AbstractRelational databases have been around for a long time, and Spatial databases have exploited this feature for close to two decades. The recent past has seen the development of NoSQL non-relational databases, which are now being adopted for spatial object storage and handling too. Moreover, this is gaining ground in the context of a rising shift towards Geospatial Web Services on both the Web and mobile platforms, primarily the user-centric services, where there is a need to improve the query response time. There is a need to assess the spatial functionalities supported by NoSQL databases and gauge the appropriateness of NoSQL databases for a full-scale Geospatial application. This thesis attempts to analyse the performance and adaptability of an existing NoSQL database(MongoDB) with its inbuilt spatial functions compared with that of a SQL database(PostgreSQL) with spatial extension(PostGIS). We compare the performance of both databases on the spatial functions they currently support on simulated as well as real-world data sets. In addition, to evaluate its utility in a GIS application we implemented a routing engine, which is a primary GIS application due to its usage of a large amount of spatial data in the form of network and topology. This implementation helps assess the usability of NoSQL databases in a spatial context against the current SQL implementation (pgRouting). For spatial queries, it was found that MongoDB achieves a gain of 25x in time performance for topological queries and about 10x for aggregate queries. In the routing application for Dijkstra's implementation over the two databases indicate that the performance is better in case of small networks with less number of queries. We expect the routing application to outperform the current server-client model of pgRouting once appropriate optimisations are incorporated into it. Based on the observations from this study, it is anticipated that NoSQL databases may be better suited for simultaneous multiple-user query systems including on Web-GIS and mobile-GIS platforms. While this study has helped to show the potential of NoSQL databases in the context of spatial queries and spatial data handling, further studies may be carried out on other spatial query functions as well as other spatial algorithms to evaluate the use or adoption of NoSQL based spatial databases. Also, performance of MongoDB for simultaneous queries over distributed systems can provide appropriate insights to its utility. Full thesis: pdf Centre for Spatial Informatics |
||||||||
Copyright © 2009 - IIIT Hyderabad. All Rights Reserved. |