Enhancing Hadoop MapReduce Performance for Scientific Data using NoSQL Database

摘要

Scientific data sets usually have similar jobs that are frequently applied to the data by different users. In addition, many of these data sets are unstructured, complex, and required fast and simple processing. In order to increase the performance of the existing Hadoop and MapReduce algorithm, it is necessary to develop an algorithm based on the type of data sets and requirements of the jobs. In this poster, we represent a Hadoop MapReduce environment that uses genomic and biological data as an example of unstructured and complex data.