Reconfigurable Computing Architectures for Edge Computing Applications

Abstract

With the explosive growth of Internet of Things (IoT) devices and edge com
puting applications, the way our data is being processed and analyzed is 
undergoing a major revolution. With the exponential growth of the volume 
of data being generated at the network edge, traditional computing archi
tectures are no longer able to cope with the strict demands of real time 
processing, low latency and energy efficiency. As a result, the reconfigurable 
computing architectures have emerged as a promising solution for edge com
puting scenarios. Dynamic allocation of hardware resources to different ap
plications is a main motivator for reconfigurable computing and could lead 
to substantial performance and energy efficiency improvements over fixed 
architectures. Reconfigurable edge architectures that bring adaptable com
puting capabilities near the data source are designed to overcome the lim
itations of cloud centric models for intelligent, responsive edge applications. 
In this comprehensive study of reconfigurable computing architecture in a 
new domain for edge computing, key technologies, design strategies, and 
application areas are addressed. In this work we analyze the characteristics 
and opportunities of the edge environment, and how reconfigurable system 
address these requirements. By synthesizing research and industry develop
ments over the past few years, we provide perspectives on the existing art 
and the promising future avenues.