BIG-BIO: - big data hadoop-based analytic cluster framework for bioinformatics

摘要

Big volume of bioinformatics data needs high processing powers. BIG-BIO is one of the solutions for addressing these challenges. BIG-BIO is a big data analyst MapReduce Hadoop Cluster for Bioinformatics applications. BIG-BIO is tested by implementing the bioinformatics wordcount problem and its applications using MapReduce programming pattern. BIG-BIO counts the number of occurrence of each word in a text and extracts unique words in molecular sequences. The application is characterized by almost low-weight computation on big size data sets. Performance of BIG-BIO is tested. BIG-BIO framework could analyze bioinformatics big data faster and more efficient. It has many bioinformatics applications while maintaining good processing capabilities, scalability and ease of maintenance by using cheap commodity. BIG-BIO reduces the processing time for parallel bioinformatics algorithms compared to legacy serial and MPI based applications. Testing BIG-BIO stated that it scales automatically with size of data. BIG-BIO can be portable on many Hadoop infrastructures without modification, with accelerating data-intensive bioinformatics analysis.