The diverse computing services offered by the cloud computing paradigm have escalated the interest in cloud deployment to a great extent. Cloud systems need to be resilient to uncertainties and perturbations. However, the perturbations in a cloud environment may cause the performance to degrade and violate the Service Level Agreements (SLAs). Therefore, it is imperative to adhere to the performance assurance by guaranteeing reliability in diverse and unexpected conditions. In our research, we focused on measuring and analyzing the robustness of a cloud based scheduling system. To mitigate the negative effects of the perturbations and uncertainties existing in the system working environment we present a robust resource allocation system. In our study, we focused on a two-step line of action: (a) measurement of robustness and (b) achieving an optimized Pareto front of the scheduler system is Cloud. To address the aforesaid challenge and fulfill the required Quality of Service (QoS), this research work employs a robustness analysis of resource allocation schemes in cloud on the basis of multiple performance parameters. Due to the high number of parameters’ comparison criterion, decision of the most robust allocation scheme is quite challenging. Therefore, a dimension reduction mechanism is employed to reduce the problem complexity. Thereafter, the resource allocation schemes are evaluated for guaranteeing the systemwide performance to ensure reliability and ascertain promising performance. The experimental results depict that the order of parameter selection in the reduction process has a significant impact on the selection of the most robust allocation scheme. The performance demands of modern computing applications have led to an exponential increase in power density of on-chip devices. Not only the operational budget of the system has increased substantially, but also the temperature has experienced an alarming increase rate. The aforementioned challenges necessitate the requirement of realizing efficient mapping methodologies to overcome the resource exploitation issue in Cloud computing. This study attempts the optimization of performance, power, and temperature of multi core systems by varying the frequency of operation of the core. Our proposed resource scheduler efficiently adheres to the optimized Pareto front to address the aforementioned challenges.