The demand for efficient and reliable cloud computing systems is increasing. However, effectively managing data workloads in edge cloud systems, especially for connected cars, can be challenging. To address this issue, we have developed a new cache management technique named VG-Prefetcher Cache that uses visibility graphs to handle time series data more effectively. Our approach involves predicting future data and prefetching it into the cache, which reduces retrieval time and improves system performance. VG-Prefetcher Cache presents a promising approach for overcoming challenges in managing data workloads, thus paving the way for a more efficient and reliable cloud computing system.

International audience

During the last decade, big data management has attracted increasing interest from both the industrial and academic communities. In parallel, Cyber Security has become mandatory due to various and more intensive threats. In June 2022, a group of researchers has met to reflect on their community’s impacts on current research challenges. In particular, they have considered four dimensions: (1) dedicated systems being data processing and analytic platforms or time series management systems; (2) graphs analytics and distributed computation; (3) privacy; and (4) new hardware.

We presented ModulAr Semantic CAching fRAmework (MASCARA) that deployed Semantic Caching (SC) to perform a fast query processing based on Field Programmable Gate Arrays (FPGAs) accelerators. In addition of the accelerators, cache management plays an important role to address coalescing strategy and replacement policy so as to maximize the performance of FPGA caching. Therefore, in this paper, we present a coalescing heuristic with a new replacement function that leverages advantages of traditional strategies and overcomes their drawbacks. The proposed heuristic reduces response time, improves data availability, and saves cache space with respect to the semantic locality of query workload.

International audience

Query optimization is a challenging process of DBMSs. When tackling query optimization in the cloud, there exists a simultaneous need of providing an optimal physical query execution plan, as well as an optimal resource configuration among available ones. Cloud computing features like resource elasticity and pricing make the process of finding this optimal query plan a multi-objective problem, with the monetary cost being an equally important factor to query execution time. Apache Spark is a popular choice for managing big data in the cloud. However, query optimization in its SQL module (Spark SQL) involves a number of limitations due to the rule-based nature of its optimizer, Catalyst. We propose a multi-objective cost model for the extension of the query optimizer of Apache Spark, aiming to minimize both objectives of query execution time and monetary cost, as well as a methodology for exploring the space of Pareto-optimal query plans and selecting one. The cost model is implemented and tuned, and an experimental study is conducted to validate its accuracy.

There are fewer female authors than male authors in the field of scientific research. However, there is not yet a system that provides a way to analyze the data that is available, and to backup that claim. This paper illustrates the upgrade of a tool previously made, in order to make it more efficient and add new features. Such new features are the keywords cloud or the new statistical functionality. Sources, references and other information on the article will be displayed for each articles retrieved. Genders of the authors will be determined using a database linking first names to genders, to be able to get accurate statistics on a large number of gathered articles.

In this article we introduce a Data and Knowledge Integration Model and a Collaborative Platform for fact-oriented Agricultural Biodiversity Management that is inspired by the conservation and sustainable use of biodiversity within agricultural landscapes, which is essential for the future of agriculture and food security. We demonstrate and validate our proposal in a realistic case study that was carried out with stakeholders from educational institutes including several government agencies from five Ministries, i.e. Ministry of Agricultural and Cooperative, Ministry of Natural Resources and Environment, Ministry of Public Health, Ministry of Commerce and Ministry of Higher Education, Science, Research and Innovation. Key challenges are how to make data inter-operation across these agencies when re-engineering the existing information system and how to make trustworthy platform for data collecting, integrating and sharing, especially, how to keep these agencies engaged throughout the project. The resulting Syntax-Semantic-Organizational Inter-operability model was proposed to provide a candidate best practice for engineering data and knowledge integration through a community-shared and reusable Data Reference Model. The resulting Data Governance Implementation across government agencies, by using BIO-AGRI-WATCH as a case study, has significant consequences regarding communication and engagement with stakeholders and dedicated team for increasing their trust in digital data sharing platform.

International audience

The use of Field Programmable Gate Arrays (FPGA) has become attractive in recent years to accelerate database analysis. Meanwhile, Semantic Caching (SC) is a technique for optimizing the evaluation of database queries by exploiting the knowledge and resources contained in the queries themselves. Organizing SC on FPGA is relevant in terms of response time and quality of results to increase system performance. To make SC scalable on FPGAs, we have proposed a ModulAr Semantic CAching fRAmework (MASCARA) in which relevant stages or modules could be convertible as accelerators on FPGAs. Therefore, in this paper, we aim to present a complementary query processing platform based on the cooperation model between MASCARA and FPGA. This novel approach extends the advantage of the classical SC, which is mainly based on Central Processing Unit (CPU), by offloading computationally intensive phases to FPGA. Moreover, MASCARA-FPGA presents the workflow of query rewriting and partial query execution in a pipelined execution model where multiple accelerators can run in parallel. In our experiments, the Query Trimming can reduce the response time by up to 3.96 times with only one accelerator used.

International audience

Cloud computing provides computing resources with elasticity following a pay-as-you-go model. This raises Multi-Objective Optimization Problems (MOOP), in particular to find Query Execution Plans (QEPs) with respect to users' preferences being for example response time, money, quality, etc. In such a context, MOOP may generate Pareto-optimal front with high complexity. Pareto-dominated based Multi-objective Evolutionary Algorithms (MOEA) are often used as an alternative solution, like Non-dominated Sorting Genetic Algorithms (NSGAs) that provide better computational complexity. This paper presents NSGA-G, a NSGA based on Grid Partitioning for improving complexity and quality of current NSGAs. Experiments on DTLZ test problems using Generational Distance (GD), Inverted Generational Distance (IGD) and Maximum Pareto Front Error prove the relevance of our solution.

Most of the existing database query optimization techniques are designed to target traditional database systems with one-dimensional optimization objectives. These techniques usually aim to reduce either the query response time or the I/O cost of a query. Evidently, these optimization algorithms are not suitable for cloud database systems because they are provided to users as on-demand services which charge for their usage. In this case, users will take both query response time and monetary cost paid to the cloud service providers into consideration for selecting a database system product. Thus, query optimization for cloud database systems needs to target reducing monetary cost in addition to query response time. This means that query optimization has multiple objectives which are more challenging than one-dimensional objectives found in traditional paradigms. Similar problems exist when incorporating query re-optimization into the query execution process to obtain more accurate, multi-objective cost estimates. This paper presents a query optimization method that achieves two goals: 1) identifying a query execution plan that satisfies the multiple objectives provided by the user and 2) reducing the costs of running the query execution plan by performing adaptive query re-optimization during query execution. The experimental results show that the proposed method can save either the time cost or the monetary cost based on the type of queries.

When running queries on a database, choosing an optimal query execution plan to minimize query costs is crucial for the query optimizer. This is especially true in mobile-cloud database systems, where there are multiple costs to execute a query plan such as money, time and energy. In order to fulfill different cost objectives for different users, some query optimizers allow users to select the query execution plan from a Pareto Set based on Skyline queries. The users must select from a potentially large quantity of options, and these options present the values of costs. It is not straightforward to the users how to compare these values in such a way to choose the option that suits their needs best. This increases the possibility for users to choose in-optimal options, and the amount of time spent to make that choice. However, the existing user interaction model during multi-objective query processing is unable to solve this issue. To fill this gap, this paper presents a new user interaction model in multi-objective query processing. This model introduces the administrators, or super users, to the user interaction process, allowing them to preset Weight Profiles and their logical descriptions. Weight Profiles contain objective preferences for the users before the query is executed. By using this model, the users can select a Weight Profile that will obtain their optimal query execution plan, and the process of choosing will be more accurate and efficient.

Recently, grids and pervasive systems have been drawing increasing attention in order to coordinate large-scale resources enabling access to small and smart devices. In this paper, we propose a caching approach enabling to improve querying on pervasive grids. Our proposal, called semantic pervasive dual cache, follows a semantics oriented approach. It is based on the one hand on a clear separation between the analysis and the evaluation process, and on the other hand on the cooperation between client caches considering light analysis and proxy caches including evaluation capabilities. Such an approach helps load balancing, making the system more scalable. We have validated semantic pervasive dual cache using analytic models and simulations. Results obtained show that our approach is quite promising.