This research aims to find an optimal configuration regarding power consumption, together with other desired measures such as reliability and development cost. This purpose is achieved by searching through the feature space which is defined as the set of all feasible SPL configurations. Combining the multi-modal data is one of the main challenges of this research. Another main challenge is the nature of the cloud including loads and usage trends which dynamically changes over time.

The proposed method is based on designing a mathematical model to learn a mapping from the features of the software to the power consumption. The target system determines that whether a specific configuration is acceptable from the power consumption point of view and therefore it can be seen as a solution to a classification problem. The main steps of the proposed method are as follows. First, we create a dataset for cloud systems containing the software configurations as the features of training data and power consumption of the corresponding system as the desired output. Second, we extract the set of meaningful and independent features by considering the causal network. Third, we employ representation learning \cite{Bengio_2013} for creating a novel feature space. We aim to investigate the benefits of using deep neural networks in this application. After these steps, we design and train a classifier for the training data set by considering the causal relations as a priori.