摘要 Abstract 

Cloud environments enabled with minimum upfront capital investment and maximum scala-bility features allow end users to expand and shrink their demand for resources dynamically over time. However, the fluctuations in the resource demands and pre-defined size of virtual machines (VMs), and sharing of common physical machines among multiple users lead to resource wastage, excessive power consumption, increased security breaches, hampered data privacy, and performance degradation. To address these pivotal and complex challeng-ing issues, this presentation will discuss the following key contributions including an Evolu-tionary Quantum Neural Network (EQNN) model towards prediction of a dynamic and exten-sive range of cloud workloads, and Quantum Machine learning-driven Malicious User Pre-diction (QM-MUP) model that estimates the vicious entity present in the communication sys-tem precedently before data allocation by scrutinizing the behavior of each user and esti-mating the probable data. Both models are the ingenious collaboration of the computational efficiency of Quantum mechanics and adaptive machine learning capabilities of evolutionary neural networks. To deal with the security challenges during data communications among multiple users, the key approach and contributions of Machine Learning and Probabilistic Analysis based data security model will be discussed. Additionally, some glimpses of secure load distribution and execution, and fault-tolerant with sustainable resource distribution will be discussed. These models illustrate effective cloud resource management; data protection through privacy-preserving data storage and analysis, secure sharing, and identification of guilty entities against data leakage in the cloud environment. All these works have been im-plemented and evaluated using a wide variety of benchmark cloud workload datasets. The achieved results and comparison with the state-of-the-art approaches validated the influen-tial performance and potency of the proposed models.