Information retrieval using the near function Selects the touristic attractions (geometry: points) found in the districts (geometry: polygons) that intersect the protected area of Arenal Information retrieval using the functions within and intersect The query specifically selects the districts that intersect the protected area of Arenal. The intersect function returns a geometry that represents the intersection of the set of points of the geometries. Information retrieval using the intersect function between two geometries of the type polygon The aim of including conditions in the where clause in an SQL query is to filter those tuples that meet certain characteristics represented by those conditions. Selects all the attributes of the San Carlos districts data set (geometry: polygon). Information retrieval with condition on clause where Specifically, two towns are selected (geometry: points) that are within the district of Florencia (geometry: polygon). That is, it takes two geometries as entry parameters and returns the number one if the first geometry is within the second one. This approach improves performance greatly. First, it determines the delimiting square of the geometry by using the R tree index, then it loads the geometry. This is a geographic query that uses the function within. Information retrieval using the function within Selects all the attributes of the towns (type of geographic data: points). The results obtained allow us to conclude that the type of query influences the response time of both DBMS. In this work, eleven types of queries were analyzed, using or not the geographical index. Here, this analysis is improved by carrying out a statistical analysis that uses real spatial data corresponding to the Huetar Norte Region of Costa Rica, focusing only in evaluating the query operation and both DBMS of interest. Finally, published the basis for the present article, with the restriction that its analysis was based on descriptive statistics. Īnother work analyzes geographical queries using real data, but it focuses on analyzing PostgreSQL, MongoDB and Neo4j. However, although the synthetic data usually favor the evaluation of queries regarding numeric and string characters, the synthetic geometries generated differ from real-world data, because the randomness of the generator can produce too-square polygons and unnatural landforms. The study of also focuses on comparing geographical queries using synthetic data. Among these works, there are and, which focus on comparing the performance in the operations insert, select, update and delete. One of the main motivations that led us to conduct this research was the scarce number of works comparing PostgreSQL’s and MongoDB’s performance. Lastly, both are open code DBMS and Geoserver gives them support because in its version 2.11.4, Geoserver included a data connexion and publication component from MongoDB. We chose MongoDB because, to date, it was the only document-based NoSQL database that supports line intersection and point containment queries. On the other hand, there are currently over 225 NoSQL databases with only a reduced number supporting geographical data operations, among which Neo4j, CouchDB, MongoDB and ArangoDB outstand in this area. PostgreSQL’s extension, PostGIS,, is highly optimized for spatial queries and its large quantity of spatial functions make it relevant for this research project. Since PostgreSQL was one of the first databases to address spatial issues, we selected it to construct the hybrid and distributed database. The aim was to choose appropriately the DBMS (Data Base Management System) to be employed by the Web Map Service (WMS) implemented by Geoserver. However, while developing the project, it became necessary to establish if there was any difference in the response times of PostgreSQL and MongoDB, according to the type of query and use of geographical indexes. Ĭurrently, a hybrid and distributed database that uses PostgreSQL and MongoDB database engines is being developed for future implementation in the Geoserver base architecture to evaluate whether this modification enhances its performance. Hybrid databases work as an abstraction layer over SQL and NoSQL databases. Īs a response to the issue raised above, work has been conducted on producing hybrids between the relational (SQL) and non-relational (NoSQL) database paradigms. While using it, users also participate actively in the generation of geographical data, in the sense that most people have currently become a mobile sensor that registers and records large volumes of data that require higher computing capability and more advanced and efficient processing and analyzing methods. The demand for geographical information has grown considerably in recent years and citizens are among the main generators of this geographical explosion.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |