An Analysis of the Performance and Configuration Features of MySQL Document Store and Elasticsearch as an Alternative Backend in a Data Replication Solution

In recent years, with the increase in the volume and complexity of data, choosing a suitable database for storing huge amounts of data is not easy, because it must consider aspects such as manageability, scalability, and extensibility. Nowadays, the NoSQL databases have gained immense popularity for their efficiency in managing such datasets compared to relational databases. However, relational databases also exhibit some advantages in certain circumstances, therefore many applications use a combined approach: relational and non-relational. This paper performs a comparative evaluation of two popular open-source DBMSs: MySQL Document Store and Elasticsearch as non-relational DBMSs; this comparison is based on a detailed analysis of CRUD operations for different amounts of data showing how the databases could be modeled and used in an application. A case-study application was developed for this purpose in Java programming language and Spring framework using for data storage both relational MySQL and non-relational Elasticsearch and MySQL Document Store. To model the real situation encountered in several developed applications that use both relational and non-relational databases, a data replication solution that imports data from the primary relational MySQL database into Elasticsearch and MySQL Document Store as possible alternatives for more efficient data search was proposed and implemented.

A dataset of regressions in web applications detected by end-to-end tests

End-to-end tests present many challenges in the industry. The long-running times of these tests make it unsuitable to apply research work on test case prioritization or test case selection, for instance, on them, as most works on these two problems are based on datasets of unit tests. These ones are fast to run, and time is not usually a considered criterion. This is because there is no dataset of end-to-end tests, due to the infrastructure needs for running this kind of tests, the complexity of the setup and the lack of proper characterization of the faults and their fixes. Therefore, running end-to-end tests for any research work is hard and time-consuming, and the availability of a dataset containing regression bugs, documentation and logs for these tests might foster the usage of end-to-end tests in research works. This paper presents a) a dataset for this kind of tests, including six well-documented manually injected regression bugs and their corresponding fixes in three web applications built using Java and the Spring framework; b) tools for easing the execution of these tests no matter the infrastructure; and c) a comparative study with two well-known datasets of unit tests. The comparative study shows that there are important differences between end-to-end and unit tests, such as their execution time and the amount of resources they consume, which are much higher in the end-to-end tests. End-to-end testing deserves some attention from researchers. Our dataset is a first effort toward easing the usage of end-to-end tests in research works.

UPOTREBA GRAPHQL TEHNOLOGIJE ZA RAZVOJ VEB APLIKACIJA

Ovaj rad opisuje implementaciju GraphQL tehnologije, u obliku veb aplikacije za prodaju polovnih automobila. U radu su date osnove GraphQL-a, specifikacija rešenja koja je predstavljena UML dijagramima kao i najvažniji implementacioni delovi sa klijentske i serverske strane. Klijentski deo aplikacije je rađen uz pomoć Angular okruženja i Apollo klijenta. Serverski deo je pisan u programskom jeziku Java i korišćen je Spring framework.

DESIGN AND IMPLEMENTATION OF PERSONNEL MANAGEMENT INFORMATION SYSTEM BASED ON SSM

Personnel management is a series of management activities for the company to select, use, train, evaluate, reward and punish its employees. The personnel management information system designed and implemented in this paper integrates the functions of batch import of information, instant update of personnel information, position transfer, information inquiry and statistical analysis. This system is oriented to various users such as managers at all levels of small and medium-sized enterprises and institutions. In this paper, using the aspect-oriented programming, Inversion of Control features of the Spring framework, using spring MVC model and using the ORM (Object Relational Mapping, Object Relational Mapping) feature of MyBatis, using B/S architecture and MySQL database, with high scalability, high refactorability as the design goal, designs and implements an Personnel Management Information System. The system builds a personnel database covering more than 20 information about personnel's basic information, education background, and positions, and realizes functions such as batch import of information, instant update of personnel information, job transfer, entry management, resignation management, information inquiry and statistical analysis. At the same time, it is ensured that these functions will not be illegally operated by users without authority, so that the system has higher security. This system provides an integrated solution for the informatization of personnel management and has high practical and commercial value.