The wind energy has emerged as one of the safest growing renewable energy to address the crisis witnessed in power generation. Wind speed is an important factor in wind power production and integration. However, the complex nature of wind speed limits the dependability and induces high fluctuation in power generation. The accurate prediction of wind speed energy with minimum accepted errors will increase to harness the energy content in a wind efficiently. In recent years, machine learning algorithms are used to analyze and predict data to make better decisions. Ensemble model is one of the supervised machine learning approaches to predict numerical data. In this project, the speed of wind is predicted through XGBoost 4j package in a distributed programming environment of Apache Spark. XGBOOST is a short form of Extreme Gradient Boosting tool of supervised machine learning, where the training data set xi is used to predict a target variable Yi. The results are compared with different iterations in order to minimize the uncertainties and to evaluate the efficiency and accuracy. The main findings were that Ensemble model was the most accurate method.