HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Theses

Composition sure d’API fondée sur des contrats

Anis Ahmed Nacer 1
1 COAST - Web Scale Trustworthy Collaborative Service Systems
Inria Nancy - Grand Est, LORIA - NSS - Department of Networks, Systems and Services
Abstract : Virtualisation and the exponential growth of cloud offerings have turned computing and storage capacity into commodities that can be purchased on demand. This trend has also reached the services traditionally operated to enable the deployment and commissioning of applications of all kinds. In this work, we consider middleware services that are the basic elements of a software architecture. Obviously, the use of middleware services implies making choices that must be guided by service deployment quality and cost criteria. In this respect, we are faced with two main challenges, namely the selection of the cloud services that best suit the specific needs of the architects, and the composition that consists in evaluating and selecting a set of atomic services that meet the application needs. In both cases, this selection must ensure both the operational needs of the enterprise, such as storage capacity and location of services, and the quality of the service or application deployment, such as minimising deployment costs while maintaining high performance and ensured security. This thesis work contributes to the selection and composition of cloud services. Firstly, this thesis proposes a methodology to identify key comparison elements and their relationships for the selection of cloud services. This methodology is based on real data available from service providers, benchmarks and surveys achieved with cloud architects. Secondly, this thesis proposes a methodology for service selection based on (1) resolving the difficulties associated with the inability to determine how the service's configuration parameters, also known as Non-Functional Attributes (NFA), can impact the Non-Functional Requirements (NFR) such as performance, availability and security; (2) conducting a survey to assign meaningful weights to NFA and NFR based on the architects' needs; (3) normalising the NFA values according to their data types using appropriate normalisation methods; (4) using the Ordered Weighted Averaging Aggregation operator (OWA), which is a decision support method to classify cloud services according to architects' needs; and (5) proposing a service selection algorithm that uses the normalised NFA values and the OWA decision making method. Thirdly, this thesis provides a methodology to guide architects in the selection of services plans compositions for their architectures. This methodology is based on: (1) the results of the services selection study and the analysis of the services composition process from the architects' point of view; (2) An algorithm for services composition that generates composition solutions in a reasonable time, while adapting the selection of service plans according to the architects' needs, especially in terms of cost. We validated the effectiveness of our cloud service selection methodology using actual services plans configurations and feedback from domain experts. This validation was performed for two types of services: relational database cloud services and queuing cloud services. For cloud services composition, we validated and applied it to the selection of cloud services composed of two types of services: relational database cloud services and queuing cloud services.
Document type :
Theses
Complete list of metadata

https://hal.univ-lorraine.fr/tel-03598259
Contributor : Thèses Ul Connect in order to contact the contributor
Submitted on : Friday, March 4, 2022 - 5:24:31 PM
Last modification on : Saturday, March 5, 2022 - 3:30:51 AM

File

DDOC_T_2021_0265_AHMED_NACER.p...
Files produced by the author(s)

Identifiers

  • HAL Id : tel-03598259, version 1

Citation

Anis Ahmed Nacer. Composition sure d’API fondée sur des contrats. Informatique [cs]. Université de Lorraine, 2021. Français. ⟨NNT : 2021LORR0265⟩. ⟨tel-03598259⟩

Share

Metrics

Record views

67

Files downloads

29