Artificial Intelligence for IT Operations (AIOps) has emerged as a critical paradigm for maintaining the high availability and reliability of large-scale distributed systems. As modern service architectures evolve alongside advancements in IT infrastructure, their growing complexity—marked by escalating system scales and intricate dependencies among components—has intensified challenges in operational management, particularly in addressing alert storms. Despite the proliferation of AIOps methodologies, a systematic and comprehensive survey dedicated to analyzing alert storm phenomena remains absent in the literature. To bridge this gap, this study presents a novel survey that rigorously examines alert storm identification, characterization, and summarization within AIOps-driven systems. Through a structured review, our work contributes threefold to the field: First, we synthesize foundational research on microservice architectures, incident management, and alert handling, establishing a cohesive framework for understanding AIOps’ role in mitigating operational disruptions. Second, we classify existing methodologies into four distinct categories based on their technical approaches, elucidating their strengths and limitations. Finally, we evaluate performance metrics employed in alert storm management, offering insights into their applicability and efficacy. This survey not only consolidates critical knowledge for researchers and practitioners but also highlights future directions for advancing AIOps in complex, dependency-rich environments.