A Multiple Controlled Toffoli Driven Adaptive Quantum Neural Network Model for Dynamic Workload Prediction in Cloud...

Area of Science:

• Cloud Computing
• Artificial Intelligence
• Quantum Computing

Background:

• Cloud computing faces challenges in dynamic resource scaling, load balancing, and power consumption.
• Accurate workload prediction is vital for addressing these cloud computing challenges.
• Existing workload prediction methods struggle with the high variance of dynamic cloud workloads.

Purpose of the Study:

• To introduce a novel model for accurate cloud workload prediction.
• To address the limitations of current approaches in handling volatile cloud workloads.
• To optimize exploration, adaptation, and exploitation proficiencies through quantum learning.

Main Methods:

• A novel Multiple Controlled Toffoli-driven Adaptive Quantum Neural Network (MCT-AQNN) model is presented.
• Quantum computing's adaptability is integrated with machine learning algorithms.
• Multiple Controlled Toffoli (MCT) gates are utilized in the Quantum Neural Network (QNN) hidden and output layers.
• A Uniformly Adaptive Quantum Machine Learning (UAQL) algorithm is developed for training the QNN.

Main Results:

• The MCT-AQNN model demonstrates superior performance in workload prediction.
• Experiments conducted on four real-world benchmark datasets show significant accuracy improvements.
• The proposed model achieves up to 32%-96% higher accuracy compared to state-of-the-art methods.

Conclusions:

• The MCT-AQNN model offers an effective solution for complex and elastic cloud workload prediction.
• Quantum-enhanced machine learning provides more precise correlations from dynamic workloads.
• The novel approach enhances learning capabilities and prediction accuracy in cloud environments.