Search

Software testing enables development teams to maintain the quality of a software system while it evolves. The database manipulation code requires special attention in this context. However, it is often neglected and suffers from software maintenance problems. In this paper, we study the current state-of-the-practice in testing database manipulation code. We first analysed the tests of 72 open-source projects to gain insight into the coverage of database access code. The database was poorly tested: 46% of the projects did not cover with tests half of their database access methods, and 33% did not cover the database code at all. This poor coverage motivated us to study developers' challenges and best practices. (i) First, we analysed 532 questions on Stack Exchange sites and deduced a taxonomy of issues. Developers mostly looked for general best practices to test database access code. Their technical questions were related to database management, mocking, parallelisation, or framework/tool usage. (ii) Next, we examined the answers to these questions. We manually labelled 598 answers to 255 questions. We distinguished 363 solutions and organised them in a taxonomy of best practices. Most of the suggestions considered the testing environment and recommended various tools or configurations. The second largest category was database management, where many addressed database initialisation and clean-up between tests. Other categories pertained to code structure or design, concepts, performance, processes, test characteristics, test code, and mocking. We illustrate the two taxonomies through intriguing examples.(c) 2022 The Author(s). Published by Elsevier Ltd.

International audience; Underground Pumped Hydro-Energy Storage stations are sustainable options to enhance storage capacity and thus the flexibility of energy systems. Efficient management of such units requires high-performance optimization algorithms able to find solutions in a very restricted timing to comply with the responsive energy markets. In this context, parallel computing offers a valuable solution to ensure appropriate decisions that maximize the profit of the station operator, while guaranteeing the safety of the energy network. This study investigates the use of three existing algorithms in Parallel Bayesian Optimization, namely q-EGO, BSP-EGO and TuRBO. The three algorithms have different inherent behaviors in terms of parallel potential and, even though TuRBO scales better, q-EGO remains the best choice regarding the final outcomes for all investigated batch sizes and manages to get up to 5 times more profits than other approaches.

Data-intensive system evolution is a complex and error-prone process, as most evolution scenarios impact several interdependent artefacts such as the application code, the data structures or data instances. This process is becoming even more challenging with the emergence of heterogeneous databasearchitectures, commonly called hybrid polystores, that rely on a combination of several, possibly overlapping relational and NoSQL databases. This paper presents HyDRa, a framework aiming to facilitate the evolution of polystores thanks to automatically generated data access APIs. For a given polystore, aconceptual API can be derived from the conceptual schema of the polystore and its correspondences with the physical schemas of the underlying databases. Applications built on top of the generated API are then protected from future schema and data reconfiguration changes applied to the polystore. Furthermore,HyDRa automatically enforces cross-database data integrity constraints and does not require developers to master multiple data models and query languages. This paper presents HyDRa and demonstrates its main features based on open-source datasets and realistic use cases.

Managing data-intensive systems has long been recognized as an expensive and error-prone process. This is mainly due to the often implicit consistency relationships that hold between applications and their database. As new technologies emerged for specialized purposes (e.g., graph databases, document stores), the joint use of database models has also become popular. There are undeniable benefits of such multi-database models where developers combine various technologies. However, the side effects on design, querying, and maintenance are not well-known yet. In this paper, we study multi-database models in software systems by mining major open-source repositories. We consider four years of history, from 2017 to 2020, of a total number of 40,609 projects with databases. Our results confirm the emergence of hybrid data-intensive systems as we found (multi-) database models (e.g., relational and non-relational) used together in 16% of all database-dependent projects. One percent of the systems added, deleted, or changed a database during the four years. The majority (62%) of these systems had a single database before becoming hybrid, and another significant part (19%) became “mono-database” after initially using multiple databases. We examine the evolution of these systems to understand the rationale of the design choices of the developers. Our study aims to guide future research towards new challenges posed by those emerging data management architectures.

Software testing enable development teams to maintain the quality of a software system while it evolves. The database manipulation code requires special attention in this context. However, it is often neglected and suffers from software maintenance problems. In this paper, we investigate the current state-of-the-practice in testing database manipulation code. We first analyse the code of 72 projects mined from Libraries.io to get an impression of the test coverage for database code. We confirm that the database is poorly tested: 46% of the projects did not cover with tests half of their database access methods, and 33% of the projects did not cover the database code at all. To understand the difficulties in testing database code, we analysed 532 questions on StackExchange sites and deduced a taxonomy. We found that developers mostly look for insights on general best practices to test database access code. They also have more technical questions related to DB handling, mocking, parallelisation or framework/tool usage. This investigation lays the basis for future research on improving database code testing.

An increasing number of organisations rely on NoSQL technologies to manage their mission-critical data. However, those technologies were not intended to replace relational database management systems, but rather to complement them. Hence the recent emergence of heterogeneous database architectures, commonly called hybrid polystores, that rely on a combination of several, possibly overlapping relational and NoSQL databases. Unfortunately, there is still a lack of models, methods and tools for data modeling and manipulation in such architectures. With the aim to fill this gap, we introduce HyDRa, a conceptual framework to design and manipulate hybrid polystores. We present the HyDRa textual modeling language allowing one to specify (1) the conceptual schema of a polystore, (2) the physical schemas of each of its databases, and (3) a set of mapping rules to express possibly complex correspondences between the conceptual schema elements and the physical databases.

Database schema change has long been recognized as a complex, time-consuming and risky process. It requires not only the modification of database structures and contents, but also the joint evolution of related application programs. This coevolution process mainly consists in converting database queries expressed on the source database schema, into equivalent queries expressed on the target database schema. Several approaches, techniques and tools have been proposed to address this problem, by considering software systems relying on a single database. In this paper, we propose an automated approach to query adaptation for schema changes in hybrid polystores, i.e., data-intensive systems relying on several, possibly heterogeneous, databases. The proposed approach takes advantage of a conceptual modeling language for representing the polystore schema, and considers a generic query language for expressing queries on top of this schema. Given a source polystore schema, a set of input queries and a list of schema change operators, our approach (1) identifies those input queries that cannot be transformed into equivalent queries expressed on the target schema, (2) automatically transforms those input queries that can be adapted to the target schema, and (3) generates warnings for those output queries requiring further manual inspection.

International audience; Surrogate-based optimization is widely used to deal with long-running black-box simulation-based objective functions. Actually, the use of a surrogate model such as Kriging or Artificial Neural Network allows to reduce the number of calls to the CPU time-intensive simulator. Bayesian optimization uses the ability of surrogates to provide useful information to help guiding effectively the optimization process. In this paper, the Efficient Global Optimization (EGO) reference framework is challenged by a Bayesian Neural Network-assisted Genetic Algorithm, namely BNN-GA. The Bayesian Neural Network (BNN) surrogate is chosen for its ability to provide an uncertainty measure of the prediction that allows to compute the Expected Improvement of a candidate solution in order to improve the exploration of the objective space. BNN is also more reliable than Kriging models for high-dimensional problems and faster to set up thanks to its incremental training. In addition, we propose a batch-based approach for the parallelization of BNN-GA that is challenged by a parallel version of EGO, called q-EGO. Parallel computing is a highly important complementary way (to surrogates) to deal with the computational burden of simulation-based optimization. The comparison of the two parallel approaches is experimentally performed through several benchmark functions and two real-world problems within the scope of Tuberculosis Transmission Control (TBTC). The study presented in this paper proves that parallel batched BNN-GA is a viable alternative to q-EGO approaches being more suitable for high-dimensional problems, parallelization impact, bigger databases and moderate search budgets. Moreover, a significant improvement of the solutions is obtained for the two TBTC problems tackled.

Relational database management systems (RDBMS) andNoSQL database systems have been built in order to meetdierent requirements. Organisations are therefore increasinglyusing hybrid data persistence architectures, that wecall hybrid database systems or hybrid polystores. The evolutionof such systems is more complex as NoSQL technologiesimply new challenges such as handling data heterogeneity.Existing work on NoSQL database evolution exposes thedierent problems regarding data heterogeneity and mainlyprovide technology-specic solutions. In this PhD researchwe propose to apply a schema engineering approach to handlehybrid database system evolution at a conceptual level.Our starting point is an existing unied conceptual datamodel for hybrid databases. Our thesis objectives include(1) to enrich this conceptual data model with generic schemaevolution operators, (2) to specify the semantics of thoseoperators depending on the underlying data models, (3) todesign a generic evolution framework and (4) to develop itsproof-of-concept implementation. Our approach will help topropagate conceptual schema changes to all impacted softwareartefacts, namely native data structures, database contents,queries and programs.

International audience; This paper presents a comparison between two surrogate-assisted optimization methods dealing with two-stage stochastic programming. The Efficient Global Optimization (EGO) framework is challenging a method coupling Genetic Algorithm (GA) and offline-learnt kriging model for the lower stage optimization. The objective is to prove the good behavior of bayesian optimization (and in particular EGO) applied to a real-world two-stage problem with strong dependencies between the stages. The problem consists in determining the optimal strategy of an electricity market player participating in reserve (first stage) as well as day-ahead energy and real-time markets (second stage). The decisions optimized at the first stage induce constraints on the second stage so that both stages can not be dissociated. One additional difficulty is the stochastic aspect due to uncertainties of several parameters (e.g. renewable energy-based generation) that requires more computational power to be handled. Surrogate models are introduced to deal with that additional computational burden. Experiments show that the EGO-based approach gives better results than GA with offline kriging model using smaller budget.

The overarching objective underlying this research is to develop an online tool, based on a parallel corpus of French Belgian Sign Language (LSFB) and written Belgian French. This tool is aimed to assist various set of tasks related to the comparison of LSFB and French, to the benefit of general users as well as teachers in bilingual schools, translators and interpreters, as well as linguists. These tasks include (1) the comprehension of LSFB or French texts, (2) the production of LSFB or French texts, (3) the translation between LSFB and French in both directions and (4) the contrastive analysis of these languages. The first step of investigation aims at creating an unidirectional French-LSFB concordancer, able to align a one- or multiple-word expression from the French translated text with its corresponding expressions in the videotaped LSFB productions. We aim at testing the efficiency of this concordancer for the extraction of a dictionary of meanings in context. In this paper, we will present the modelling of the different data sources at our disposal and specifically the way they interact with one another.

Database reverse engineering (DRE) has traditionally been carried out by considering three main information sources: (1) the database schema, (2) the stored data, and (3) the application programs. Not all of these information sources are always available, or of sufficient quality to inform the DRE process. For example, getting access to real-world data is often extremely problematic for information systems that maintain private data. In recent years, the analysis of the evolution history of software programs have gained an increasing role in reverse engineering in general, but comparatively little such research has been carried out in the context of database reverse engineering. The goal of this paper is to contribute to narrowing this gap and exploring the use of the database evolution history as an additional information source to aid database schema reverse engineering. We present a tool-supported method for analyzing the evolution history of legacy databases, and we report on a large-scale case study of reverse engineering a complex information system and curate it as a benchmark for future research efforts within the community. We present a tool-supported method to analyze the history of a database schema.The method makes use of mining software repositories (MSR) techniques.We report on the application of the method to a large-scale case study.

Software repositories can provide valuable information for facilitating software reengineering efforts. In recent years, many researchers have started to follow a holistic approach, considering diverse software artifacts and the links existing between them. However, when analyzing data-intensive systems, comparatively little attention has been devoted to the analysis of an important system artifact: the database. Even fewer approaches attempt to uncover facts about the evolution history of database schemas. We have developed a tool-supported method for analyzing and visualizing database schema history. This paper reports early results of applying and validating this method. We discuss our experiences to date and point out several novel research perspectives in this domain.