The exponential growth of data in recent times has lead to the development of platforms like Hadoop. The Hadoop distributed file system is an important part of the Hadoop ecosystem and gives high throughput if the system is optimized to handle the problems of robustness, reliability and validity. Researchers have explored a limited number of parameters to optimize the performance of Hadoop clusters. However, there has been little previous work investigating the role of network traffic information for the optimization of such a system. Our current work explores the development of a robust methodology to measure the performance of a HDFS system. This will involve finding the factors affecting the measurements, identification of confounding factors and the set of optimal parameters to enhance the performance of a Hadoop-based software defined network environment. We will evaluate our implementation by deploying bioinformatics problem in our proposed optimized environment.