| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 4.7 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
Processos de entrega de software tradicionais não são eficientes para uma empresa que produz software se manter competitiva. A crescente necessidade de entregar software em ciclos cada vez mais curtos fomentou o crescimento de abordagens e tecnologias que permitam automatizar este tipo de processos. No entanto, estas abordagens apresentam limitações quando a qualidade de entrega do software depende da qualidade dos dados que o mesmo produz. Com isso em mente, o principal objetivo desta dissertação consiste em adotar um mecanismo de entrega de software automático, eficaz, eficiente e fiável dentro da proGrow S.A. de forma a que o seu principal negócio, que é a geração de informação a partir de dados recolhidos, seja feito da forma mais eficiente e fiável possível. Após a análise detalhada do estado de arte atual, foi elaborada uma solução que implementa uma pipeline de implantação que incluí uma estratégia de testes que desafia o paradigma de testes tradicional, projetada para o contexto das soluções desenvolvidas. Deste modo, a solução é capaz de entregar software automaticamente, garantindo a sua qualidade através de uma estratégia de testes robusta, capaz de validar a qualidade da aplicação através dos seus dados. Assim, a solução desenvolvida é capaz de reduzir o tempo de desenvolvimento de uma nova versão, bem como garantir que o processo de escrutínio para deteção de erros nos dados gerados é feito de forma automática e repetível, com o objetivo de criar um padrão de qualidade nos dados gerados.
Traditional software delivery processes are not efficient for a company that produces software to remain competitive. The growing need to deliver software in ever shorter cycles has fostered the growth of approaches and technologies that make it possible to automate this type of process. However, these approaches have limitations when the quality of software delivery depends on the quality of the data it produces. With this in mind, the main goal of this dissertation is to adopt an automatic, effective, efficient and reliable software delivery mechanism within proGrow S.A. so that its core business, which is generating information from collected data, is done in the most efficient and reliable way possible. After a detailed analysis of the current state of the art, a solution was devised that implements a deployment pipeline that includes a testing strategy that challenges the traditional testing paradigm, designed for the context of the solutions developed. In this way, the solution is able to deliver software automatically, guaranteeing its quality through a robust testing strategy, capable of validating the quality of the application through its data. Thus, the solution developed is able to reduce the development time of a new version, as well as ensuring that the scrutiny process for detecting errors in the data generated is done automatically and repeatably, with the aim of creating a quality standard in the data generated.
Traditional software delivery processes are not efficient for a company that produces software to remain competitive. The growing need to deliver software in ever shorter cycles has fostered the growth of approaches and technologies that make it possible to automate this type of process. However, these approaches have limitations when the quality of software delivery depends on the quality of the data it produces. With this in mind, the main goal of this dissertation is to adopt an automatic, effective, efficient and reliable software delivery mechanism within proGrow S.A. so that its core business, which is generating information from collected data, is done in the most efficient and reliable way possible. After a detailed analysis of the current state of the art, a solution was devised that implements a deployment pipeline that includes a testing strategy that challenges the traditional testing paradigm, designed for the context of the solutions developed. In this way, the solution is able to deliver software automatically, guaranteeing its quality through a robust testing strategy, capable of validating the quality of the application through its data. Thus, the solution developed is able to reduce the development time of a new version, as well as ensuring that the scrutiny process for detecting errors in the data generated is done automatically and repeatably, with the aim of creating a quality standard in the data generated.
Description
Keywords
Entrega contínua Integração contínua Automatização Testes Qualidade Continuous delivery Implantação Pipeline de implantação Continuous integration Automation Tests Quality Deploy Deployment pipeline