Non-functional properties, such as energy, time, and security (ETS) are becoming increasingly important in Cyber-Physical Systems (CPS) programming. This article describes TeamPlay, a research project funded under the EU Horizon 2020 programme between January 2018 and June 2021. TeamPlay aimed to provide the system designer with a toolchain for developing embedded applications where ETS properties are first-class citizens, allowing the developer to reflect directly on energy, time and security properties at the source code level. In this paper we give an overview of the TeamPlay methodology, introduce the challenges and solutions of our approach and summarise the results achieved. Overall, applying our TeamPlay methodology led to an improvement of up to 18% performance and 52% energy usage over traditional approaches.

In this paper, we investigate the application of early-exit strategies to quantized neural networks with binarized weights, mapped to low-cost FPGA SoC devices. The increasing complexity of network models means that hardware reuse and heterogeneous execution are needed and this opens the opportunity to evaluate the prediction confidence level early on. We apply the early-exit strategy to a network model suitable for ImageNet classification that combines weights with floating-point and binary arithmetic precision. The experiments show an improvement in inferred speed of around 20% using an early-exit network, compared with using a single primary neural network, with a negligible accuracy drop of 1.56%.

Graph processing is an area that has received significant attention in recent years due to the substantial expansion in industries relying on data analytics. Alongside the vital role of finding relations in social networks, graph processing is also widely used in transportation to find optimal routes and biological networks to analyse sequences. The main bottleneck in graph processing is irregular memory accesses rather than computation intensity. Since computational intensity is not a driving factor, we propose a method to perform graph processing at the edge more efficiently. We believe current cloud computing solutions are still very costly and have latency issues. The results demonstrate the benefits of a dedicated sparse graph processing algorithm compared with dense graph processing when analysing data with low density. As graph datasets grow exponentially, traversal algorithms such as breadth-first search (BFS), fundamental to many graph processing applications and metrics, become more costly to compute. Our work focuses on reviewing other implementations of breadth-first search algorithms designed for low power systems and proposing our solution that utilises advanced enhancements to achieve a significant performance boost up to 9.2x better performance in terms of MTEPS compared to other state-of-the-art solutions with a power usage of 2.32W.

The trend for heterogeneous embedded systems is the integration of accelerators and general-purpose CPU cores on the same die. In these integrated architectures, like the Zynq UltraScale+ board (CPU+FPGA) that we target in this work, hardware support for shared memory and low-overhead synchronization between the accelerator and the CPU cores make the case for exploring strategies that exploit a tight collaboration between the CPUs and the accelerator. In this paper we propose a novel lightweight scheduling strategy, FastFit, targeted to FPGA accelerators, and a new scheduler based on it, named MultiFastFit, which asynchronously tackles heterogeneous systems comprised of a variety of CPU cores and FPGA IPs. Our strategy significantly reduces the overhead to automatically compute the near-optimal chunksizes when compared to a previous state-of-the-art auto-tuned approach, which makes our approach more suitable for fine-grained applications. Additionally, our scheduler MultiFastFit has been designed to enable the efficient co-execution of work among compute devices in such a way that all the devices are busy while minimizing the load unbalance. Our approaches have been evaluated using four benchmarks carefully tuned for the low-power UltraScale+ platform. Our experiments demonstrate that the FastFit strategy always finds the near-optimal FPGA chunksize for any device configuration at a reasonable cost, even for fine-grained and irregular applications, and that heterogeneous CPU+FPGA co-executions that exploit all the compute devices are usually faster and more energy efficient than the CPU-only and FPGA-only executions. We have also compared MultiFastFit with other state-of-the-art scheduling strategies, finding that it outperforms other auto-tuned approach up to 2x and it achieves similar results to manually-tuned schedulers without requiring an offline search of the ideal CPU-FPGA partition or FPGA chunk granularity.

Heterogeneous processors, formed by binary compatible CPU cores with different microarchitectures, enable energy reductions by better matching processing capabilities and software application requirements. This new hardware platform requires novel techniques to manage power and energy to fully utilize its capabilities, particularly regarding the mapping of workloads to appropriate cores. In this paper we validate relevant published work related to power modelling for heterogeneous systems and propose a new approach for developing run-time power models that uses a hybrid set of physical predictors, performance events and CPU state information. We demonstrate the accuracy of this approach compared with the state-of-the-art and its applicability to energy aware scheduling. Our results are obtained on a commercially available platform built around the Samsung Exynos 5 Octa SoC, which features the ARM big.LITTLE heterogeneous architecture.