This study presents a comprehensive approach to cyber security attack prevention in cloud servers, leveraging optimized machine learning (ML) algorithms. Specifically, three key algorithms are employed: (i) the Optimizable Efficient Linear Algorithm, (ii) the Optimizable Kernel Algorithm, and (iii) the Optimizable Neural Networks Algorithm, noted for its high accuracy. By harnessing the capabilities of these algorithms, the research aims to fortify cloud server defences against various cyber threats. Through optimization techniques, the algorithms are tailored to enhance performance and efficacy in detecting and mitigating security breaches. The study underscores the importance of integrating advanced ML methodologies in cyber security strategies, particularly in cloud computing environments where data protection is paramount. Results demonstrate the effectiveness of the proposed approach in bolstering the resilience of cloud servers against malicious activities, offering a proactive defences mechanism against evolving cyber threats.