Data Consistency in Microservices Compositions

Financiador

Organização

Publicações

Detection of transaction consistency problems in microservices

Publication . Silva, André Filipe dos Santos; Lopes, Maria Antónia Bacelar da Costa; Fonseca, Alcides Miguel Cachulo Aguiar

The widespread adoption of microservices architecture in modern software development has introduced significant challenges in maintaining data consistency across distributed services. This thesis addresses the problem of detecting transaction consistency anomalies within microservices-based systems, which are prone to issues due to the loss of isolation in distributed transactions. To tackle this problem, we propose STRIVE; a novel approach that leverages static analysis techniques to identify potential serialization anomalies in microservices. By modeling system operations and their interactions as a graph, the algorithm detects cycles that indicate possible consistency violations. This method enables the early detection of anomalies during the development phase, thereby preventing data inconsistencies that could arise after deployment. The work includes the design of a Domain-Specific Language (DSL) to facilitate the description of microservices systems, allowing developers to model their systems at a high level of abstraction without needing low-level code. The effectiveness of STRIVE is demonstrated through case studies, showing its ability to identify potential issues in various microservices architectures. Furthermore, we also identify several areas for future exploration, including the extension of the algorithm to support additional features and the refinement of the DSL. Towards improving the reliability of microservices architectures, this thesis provides a tool that enables the proactive detection of transaction consistency issues, contributing to the broader effort of making distributed systems more robust and reliable.

2024Dissertação de mestrado

Acesso aberto

Ver mais

Entidade financiadora

Fundação para a Ciência e a Tecnologia

Programa de financiamento

3599-PPCDT

Número da atribuição

PTDC/CCI-COM/2156/2021