Today’s state-of-the-art Big data analytics engines handle masses of data, but will reach to their limits, as the future Big data flood is predicted to still grow with an increasing speed. Hence we need to think about the next development phase and future features of Big data analytics engines. In this paper, we discuss possible future enhancements in the area of Big data analytics with focus on emergent models, frameworks, and hardware technologies. We point out a selection of new challenges and open research questions.

Introduction

Big data experts characterize Big data with certain words starting with a big ‘V’. The most common ones are coined to Laney [47], which uses Volume to express that Big data must have some certain data quantity, Velocity to refer to an enormous speed of new data, and Variety to remember that in usually heterogeneous environments there are a lot of data types and formats to be processed. The community actively took over the idea of big V’s and, as a result, proposed an increasing number of V’s (e.g., [90] for 7 V’s and [19] for the top 10 list of V’s). One of these most important newly proposed big V’s is the Value, which refers to the data usefulness: We should be able to process the today’s masses of heterogeneous data in order to conclude new information, to use the data for new applications and have some benefits from them. Otherwise, handling with Big data would not be meaningful for us. Big data analytics engines are already able to handle masses of data and offer sophisticated technologies to address the volume, velocity, and variety properties. However, in order to handle the future’s increasing demand of Big data processing, we need to think about new technologies and computing paradigms to be integrated into future generations of Big data analytics engines. Furthermore, we need to discuss ways to increase the value of Big data. In this paper, we hence discuss possible next steps for future generations of Big data analytics engine. We explain and analyze the possible future integration of emergent hardware technologies into existing frameworks for Big data analytics. Furthermore, we look at new computing paradigms besides cloud computing, which may be applied in future Big data analytics engines as well. Moreover, we shortly discuss the Semantic Web offering a data model targeting at the value property of Big data. The contributions of this paper include: – an overview of the current state of big data analytics engine, emergent hardware technologies and new computing paradigms, and their relation to the Semantic Web, – a comparison of Moore’s law with the properties of top ranked parallel computers, – an analysis of high-end hardware of different types of parallel architectures, and – a selection of new challenges and open research questions in the intersection of the discussed areas for data management problems of Big data analytics engines with special focus on – Big data analytics engines for new computing paradigms like fog and dew computing, – the Semantic Internet of Things area and – index accesses of hardware-accelerated Big data analytics engines.