Cargando…
Unifying access to data from heterogeneous sources through a RESTful API using an efficient and dynamic SQL-query builder
The Web Based Monitoring (WBM) system of the Compact Muon Solenoid experiment at the CERN Large Hadron Collider was introduced in 2006. Over the last decade, this monitoring system has been extended and continuously updated with different technologies and approaches and has therefore nowadays become...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2018
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/2644673 |
Sumario: | The Web Based Monitoring (WBM) system of the Compact Muon Solenoid experiment at the CERN Large Hadron Collider was introduced in 2006. Over the last decade, this monitoring system has been extended and continuously updated with different technologies and approaches and has therefore nowadays become hard to maintain. To resolve these problems, the WBM system will be redesigned using state-of-the-art technologies to solve these issues and to prepare the system for the years of operation to come. This thesis describes the development of the first prototype as a RESTful API of the aggregation layer of this upgrade. The aggregation layer will be used to enable access to the stored data for the presentation layer of the upgrade. Before describing the work on the prototype, an overview over the topics to form the basis is given. Subsequently, the task, the required outcome and prerequisites as well as the existing system were being analyzed to formulate adequate requirements for the prototype. In addition to this analysis, an evaluation of possible technologies was performed including a proof of concept implementation for each technology. After deciding on a technology, the API and the connection to the database was designed to fit the requirements. The design is then implemented with a special focus on two services. Firstly a developed converter service which is needed to be able to set the data from the different data sources in relation to each other. Secondly a custom query builder used to dynamically create database queries based on the request made to the API. To prove the capabilities of the developed prototype, a performance test was carried out as well as a functional evaluation to show that the requirements are fulfilled. The conclusion then sums up the work done with an emphasis on occurred problems like the non-heterogeneous data sources and the challenging database layout. In addition to this, the further development of the project is addressed by making suggestions on what could be considered interesting for future work. |
---|