Chapter 12
Full Access

FPGA Solutions for Big Data Applications

First published: 17 February 2017

Abstract

Big Data analytics is the process by which value is created from the data and involves the loading, processing and analysis of large data sets. This chapter gives details on Big Data analytics and various forms of data mining. The acceleration of Big Data analytics is discussed and the concepts of scaling up and scaling out are introduced. A number of major system developments have occurred which strongly indicate the potential of FPGA technology in new forms of computing architectures and therefore Big Data applications. There have been a number of classification and regression implementations on FPGAs. Among these are a number of ANNs, including work implementing a general regression neural network (GRNN) used for iris plant and thyroid disease classification. They have developed an FPGA implementation using VHDL-based tools; it comprises summation, exponential, multiplication and division operations. The chapter also describes the computation of k-means clustering.