Resource Allocation and Service Level Agreements (SLAs) Management efficiently distribute computing resources to meet service demands while ensuring adherence to agreed SLA. This is critical for maintaining efficacy, reliability, and user satisfaction in cloud and distributed systems. However, dynamic workload variability can lead to inefficient resource utilization. Additionally, ensuring strict SLA compliance is challenging due to unpredictable system failures and latency issues. In this manuscript, Predicting Energy Efficiency in Cloud Computing Systems for Enhanced Sustainability and Service Level Agreements (CCEE-BSCNN) is proposed. At first, input data is collected from Cloud Computing Performance Metrics Dataset. To execute this, the input data is pre-processed using the Neural Correlation Integrated Adaptive Point Process Filtering (NCIAPPF), which handles missing values and performs normalization on the input data. After that, the pre-processed data is fed into the Binarized Simplicial Convolutional Neural Network (BSCNN), which is used to predict cloud computing energy efficiency. Then the proposed CCEE-BSCNN is implemented in Python and the performance metrics like Accuracy, Precision, Recall, and Specificity. The proposed CCEE-BSCNN method attains 99.06% higher Accuracy and 98.24% higher Precision, 98.36% higher Recall, and 98.49% higher Specificity when analysed through existing techniques like SHAP-Enhanced Resource Allocation for VM Scheduling and Efficient Cloud Computing (ERA-ECC-SVM), Proactive resource allocation utilizing amazon chronos driven time series model for sustainable cloud computing (PRA-ACTS-SCC-LSTM), and An effectual deep reinforcement learning driven task scheduler in cloud-fog environment (SCFE-CNN) respectively.