ANN-Based System Throughput Prediction for As-a-Service Configurations

The machine learning model is trained in advance on historical system configuration and throughput data to learn the relationship between hardware specifications and performance metrics. This preliminary training enables the model to predict throughput for new configurations without requiring manual testing, thus resolving the contradiction between accuracy and time consumption.

Instead of physically testing each system configuration to measure throughput, the patent creates a virtual model (machine learning model) that copies the behavior of actual systems. This model can predict throughput for any given configuration instantaneously, eliminating the time-consuming manual testing process while maintaining prediction accuracy.

The patent replaces the mechanical process of manual configuration testing and evaluation with an automated machine learning-based prediction system. The ML model automatically processes system specifications and predicts throughput, eliminating the need for manual intervention and reducing process complexity while maintaining or improving recommendation reliability.

The machine learning model enables the system to determine appropriate configurations and predict throughput autonomously without requiring subject matter experts. The model self-services by taking system specifications as input and automatically generating accurate throughput predictions and recommendations, thus reducing both complexity and resource requirements.