Big Data inherits dimensionality as one of the important characteristics. Dimension reduction is a complex process which aims at converting the dataset from many dimensions to a few dimensions. Dimension reduction and compression techniques are very useful to optimize the storage. In turn, it improves the performance of the cluster. This review paper aims to review different algorithms and techniques which are related to dimensionality reduction and storage encoding. This paper also provides the directions on the applicability of the suitable methodology for Big Data and distributed clusters for effective storage optimization.