Using cloud computing to process Internet of Things (IoT) application domains directly may not be the best option for all use cases, particularly those that call for quick responses. The high information bandwidth requirements of end devices are addressed by both edge computing and fog computing. There are end device requirements in both cases. Large volumes of newly created data must be processed locally under these paradigms rather than on the cloud, where it was initially created. Performance can be enhanced by taking resource management into account in cloud-based IoT installations. Workload balance, job scheduling, resource allocation, resource provisioning, and quality of service (QoS) are all included in this management. Because fog computing is growing so quickly, it offers a promising method for processing IoT data. Nevertheless, the difficulties associated with the fog landscape include scarce supply, unstable resources, a shifting fog environment, and the incapacity of accuracy. In order to manage fog-enabled cloud resources, this work provides a strategy based on the bidirectional long short-term memory (BiLSTM) and the Whale Optimization Algorithm (WOA). In order to maximize resource utilization, the BiLSTM-WOA matches requests to resources. The BiLSTM-WOA fitness function additionally accounts for latency, response time, energy usage, and bandwidth. A thorough examination revealed that the BiLSTM-WOA approach performed better in multiple parameters than other cutting-edge approaches.