Dynamic and partial reconfiguration are key differentiating capabilities of field programmable gate arrays (FPGAs). While they have been studied extensively in academic literature, they find limited use in deployed systems. We review FPGA reconfiguration, looking at architectures built for the purpose, and the properties of modern commercial architectures. We then investigate design flows and identify the key challenges in making reconfigurable FPGA systems easier to design. Finally, we look at applications where reconfiguration has found use, as well as proposing new areas where this capability places FPGAs in a unique position for adoption.

New hybrid FPGA platforms that couple processors with a reconfigurable fabric, such as the Xilinx Zynq, offer an alternative view of reconfigurable computing where software applications leverage hardware resources through the use of often reconfigured accelerators. For this to be feasible, reconfiguration overheads must be reduced so that the processor is not burdened with managing the process. We discuss partial reconfiguration (PR) on these architectures, and present an open source controller, ZyCAP, that overcomes the limitations of existing methods, offering more effective use of hardware resources in such architectures. ZyCAP combines high-throughput configuration with a high-level software interface that frees the processor from detailed PR management, making PR on the Zynq easy and efficient.

Iris is a software architecture for building highly reconfigurable radio networks. It has formed the basis for a wide range of dynamic spectrum access and cognitive radio demonstration systems presented at a number of international conferences between 2007 and 2010. These systems have been developed using heterogeneous processing platforms including general-purpose processors, field-programmable gate arrays and the Cell Broadband Engine. Focusing on runtime reconfiguration, Iris offers support for all layers of the network stack and provides a platform for the development of not only reconfigurable point-to-point radio links but complete networks of cognitive radios. This article provides an overview of Iris, presenting the unique features of the architecture and illustrating how it can be used to develop a cognitive radio testbed.

FPGAs offer high performance coupled with energy efficiency, making them extremely attractive computational resources within a cloud ecosystem. However, to achieve this integration and make them easy to program, we first need to enable users with varying expertise to easily develop cloud applications that leverage FPGAs. With the growing size of FPGAs, allocating them monolithically to users can be wasteful due to potentially low device utilization. Hence, we also need to be able to dynamically share FPGAs among multiple users. To address these concerns, we propose a methodology and a runtime system that together simplify the FPGA application development process by providing: 1) a clean abstraction with high-level APIs for easy application development; 2) a simple execution model that supports both hardware and software execution; and 3) a shared memory-model which is convenient to use for the programmers. Akin to an operating system on a computer, our lightweight runtime system enables the simultaneous execution of multiple applications by virtualizing computational resources, i.e., FPGA resources and on-board memory, and offers protection facilities to isolate applications from each other. In this paper, we illustrate how these features can be developed in a lightweight manner and quantitatively evaluate the performance overhead they introduce on a small set of applications running on our proof of concept prototype. Our results demonstrate that these features only introduce marginal performance overheads. More importantly, by sharing resources for simultaneous execution of multiple user applications, our platform improves FPGA utilization and delivers higher aggregate throughput compared to accessing the device in a time-shared manner.

Accurate air quality monitoring requires processing of multi-dimensional, multi-location sensor data, which has previously been considered in centralised machine learning models. These are often unsuitable for resource-constrained edge devices. In this article, we address this challenge by: (1) designing a novel hybrid deep learning model for hourly PM_2.5 pollutant prediction; (2) optimising the obtained model for edge devices; and (3) examining model performance running on the edge devices in terms of both accuracy and latency. The hybrid deep learning model in this work comprises a 1D Convolutional Neural Network (CNN) and a Long Short-Term Memory (LSTM) to predict hourly PM_2.5 concentration. The results show that our proposed model outperforms other deep learning models, evaluated by calculating RMSE and MAE errors. The proposed model was optimised for edge devices, the Raspberry Pi 3 Model B+ (RPi3B+) and Raspberry Pi 4 Model B (RPi4B). This optimised model reduced file size to a quarter of the original, with further size reduction achieved by implementing different post-training quantisation. In total, 8272 hourly samples were continuously fed to the edge device, with the RPi4B executing the model twice as fast as the RPi3B+ in all quantisation modes. Full-integer quantisation produced the lowest execution time, with latencies of 2.19 s and 4.73 s for RPi4B and RPi3B+, respectively.