World Modernization has led to an increase in the consumption of energy significantly over the years. Pertaining to excessive development in industrial, commercial and residential areas has increased the energy generation and usage. Understanding the energy consumption pattern is necessary as it will facilitate efficient use of energy. This understanding can be gained by interpreting energy data which can be collected using sensors installed in the smart home. Energy modelling can be helpful in assessing the amount of energy being consumed by the appliances. Integrating machine learning on acquired sensor data can result in predicting future energy consumption. In this paper we will be understanding the energy consumption pattern of a smart home where the data will be recorded using the sensors installed in the devices. A variety of models are built and tested to predict energy usage. The magnitude of data coming from multiple sensors is huge and processing this large amount of data is costly, this requires us to explore the concepts of parallelism to further optimize our prediction model in terms of performance.