Answers are grounded in this paper and cite their sources.
Microservices architecture emphasises the single-responsibility-per-service paradigm. This enables developers to focus on their assigned services, while also providing more flexibility to upgrade any service independently without affecting the entire system. This flexibility makes microservices a preferred choice for modern software systems. As adoption grows across industries, effective communication and integration between microservices become critical. This systematic literature review investigates the current academic interest in addressing microservice communication and integration challenges, with particular attention to the role of artificial intelligence (AI). Over 3,000 publications were initially retrieved from leading academic databases, from which 128 studies were selected for in-depth analysis. This review aims to answer three main research questions focusing on current issues facing microservices communication and integration and their solutions, surveying datasets and microservices benchmark systems used in reviewed papers, and lastly, applications of large language models (LLMs) in microservices-related research. The review recognises evaluated solution paradigms (including conventional and AI-enhanced approaches), and special emphasis is placed on realising emerging applications of Large Language Models (LLMs), whose capabilities show promise in optimising microservices ecosystems. This insight lays the foundation for a conceptual framework that integrates LLMs to improve microservice communication and integration. The framework suggests that LLMs can support a wide range of operational tasks, including inter-service communication analysis, resource optimisation, automated deployment, continuous monitoring, self-healing mechanisms, and test generation.
No takes yet. Share an insight, caveat, or question.
Alamer et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: