Smart Cloud Monitoring: Optimising Server Loads for Reducing Energy Consumption

The recent proliferation of cloud computing services brings forth the focal issues of energy consumption in data centers from an environmental and financial standpoint. This paper describes a monitoring system focused on energy consumption reduction in the cloud by dynamically regulating loads in servers through constrained optimization techniques. The system employs the Gated Recurrent Unit algorithm in order to analyze the patterns of server usage over time and predict future demand, thereby allowing for proactive adjustments in workload distribution. The system maintains an optimal balance between performance and energy efficiency by continuously monitoring server utilization and power consumption, taking into consideration others like actual computation abilities, temperature limits, and SLAs. Our approach intensely reduces the power consumption while keeping a similar service quality level. Our testing results have proven to work well in real-world cloud environments, thus forming a basis that will help enhance sustainable practices in the cloud.