Director: a cloud microservice selection framework

[EN] The Software Ecosystem research field has been receiving an increasing amount of attention from both academia and industry, as many organizations have been adopting them as a collaborative platform to achieve innovation faster than before. More recently, with the advent of Cloud Computing, mode...

Descripción completa

Detalles Bibliográficos
Autor: Costa, Marcelo de França
Tipo de recurso: tesis doctoral
Estado:Versión publicada
Fecha de publicación:2019
País:Brasil
Institución:Universidade Federal do Rio de Janeiro (UFRJ)
Repositorio:Repositório Institucional da UFRJ
Idioma:inglés
OAI Identifier:oai:pantheon.ufrj.br:11422/14062
Acceso en línea:http://hdl.handle.net/11422/14062
Access Level:acceso abierto
Palabra clave:Software ecosystems
Software architecture
Cloud computing
Cognitive computing
Microservices
Software engineering
CNPQ::ENGENHARIAS
Descripción
Sumario:[EN] The Software Ecosystem research field has been receiving an increasing amount of attention from both academia and industry, as many organizations have been adopting them as a collaborative platform to achieve innovation faster than before. More recently, with the advent of Cloud Computing, modern ecosystems have been offered as a service, allowing actors to contribute, but also commercialize their own solutions, by reusing available software assets, popularly in the shape of microservices, i.e., very specific functionality, usually exposed through Web technologies. With the current proliferation of platforms and microservices, an open and relevant challenge for software architects is to find and acquire the most adequate component, given a set of requirements and priorities. In this context, we propose DIRECTOR: A cloud microservice selection framework, based on complementary technical, social and semantical perspectives, i.e., by relying on objective analysis, reputation and artificial intelligence, respectively. The results obtained through a proof-of-concept (PoC), and a feasibility study conducted with industry experts, indicate that it can support software acquisition via discovery, evaluation and comparison of microservices, being able to recommend the fittest among hundreds of candidates in multiple cloud platforms.