Comparisons of Relational Databases with Big Data: a Teaching Approach

The paper's objective is to provide classification, characteristics and evaluation of available relational database systems which may be used in Big Data Predictions and Analytics. In addition, it provides a teaching approach from moving relational database to the Big Data environment. In this study, we try to answer the question of why Relational Database Bases Management Systems such as IBM’s, DB2, Oracle, and SAP fail to meet the Big Data Analytical and Prediction Requirements. The paper also compares the structured, semi-structured, and unstructured data as well as dealing with security issues related to these data formats. Finally, the operational issues such as scale, performance and availability of data by utilizing these database systems will be compared. Introduction It has been more than 40 years since the Relational Database Management Systems have been implemented and used by different organizations. This database approach was satisfying the needs of businesses dealing with static, query intensive data sets since these data sets were relatively small in nature [21, 22]. These traditional relational database systems do not answer the requirements of the increased type, volume, velocity and dynamic structure of the new data sets. So, organizations with lots of data, either have to purchase new systems or re-tool what they already have [1, 4, 5, and 8]. Big data deals with volume, velocity, variability and variety. Velocity obviously refers to how quickly the streaming data is captured [1, 6]. As more data are created and streamed the high variability and high volume as well as variety of formats are at issue [2, 5, and 7]. The incoming data from a Web log, unstructured content from the Internet, need to be captured, tagged with metadata and hierarchical file systems. As the volume, complexity, variety and velocity of digital data grow faster by the day, we need to find solutions to use these data in productive, effective, and efficient ways. With shortcomings, the ACID approach associated with the relational in the BIG Data environment, we need to effectively take advantage and manage available tools to be able to utilize the huge volume and unstructured data residing in Big Data systems [23, 25]. When it comes to data collections and utilizations, these days, one can say: “It is Brave New World”. For example, Google, Facebook, Governmental Agencies, and Financial institutions collect, transfer, manipulate, and analyze big sets of data on the daily basis [17, 21]. Creating (Collecting), Manipulating, analyzing and transferring, molecular modeling, medical images or DNA data require a newer approach of databases. Therefore, organizations need to adopt their data management practices as they load and analyze all these types of data. Whiles the traditional database approaches mainly deal with content data, new approaches call for dealing with context data [1, 11, 13, and 14]. Relational Database Management Systems (RDBMDS) RDBMS is the standard language for relational database management systems. Data is stored in the form of rows and columns where each table must have one primary key.