Cloud computing has transformed modern IT by enabling scalable, on-demand access to computing infrastructure and services. Despite its profound impact, resource allocation under varying workload conditions, SLA requirements, and cost constraints remains a central challenge. This paper reviews recent advances and methodologies in cloud resource management, emphasizing reinforcement learning and multi-agent optimization frameworks that dynamically adapt to fluctuating cloud environments. The analysis covers QoS assurance, security challenges, and cost optimization with an aim to guide future research in creating adaptive, secure, and cost-efficient cloud systems.
S et al. (Sun,) studied this question.