AI-Based Workload Prediction Models for Optimizing Serverless Resource Allocation

Serverless computing has become a popular architecture in cloud computing due to its scalability and cost-efficiency. However, unpredictable workload variations can lead to either over-provisioning or under-provisioning of resources, impacting performance and cost. This study explores the use of AI-based workload prediction models to forecast serverless workloads accurately and optimize resource allocation. The proposed model leverages machine learning techniques, including Long Short-Term Memory (LSTM) networks, to enhance predictive capabilities. Results indicate that AI-based workload forecasting reduces latency and optimizes costs compared to traditional static approaches. This paper offers a practical roadmap for serverless providers to adopt AI models for dynamic and efficient resource management.