Zoltán Ádám Mann

The advancement and widespread adoption of computing technology has yielded services that could help mitigate the climate crisis. However, the retirement of obsolete equipment, the consumption of rare earth materials, and the escalating energy demands associated with massive data processing and cloud infrastructures have raised new environmental dilemmas. Existing design and development methodologies primarily focus on fulfilling functional requirements and improving performance. In this paper, we argue that these methodologies must be augmented with sustainability considerations encompassing energy efficiency, material usage, longevity, and upgradability. Solutions at different layers of the system stack, from the physical to the application layer, must be integrated. Moreover, there should be a strong focus on the transparency of sustainability metrics across the whole computing continuum. Building on fruitful discussions at the international Lorentz workshop on Future Computing for Digital Infrastructures, we advocate novel approaches in the design, development, and operation of the computing continuum.

Applying services computing to neural networks, a service provider may provide inference with a pre-trained neural network as a service. Clients use the service to get the neural network's output on their input. To protect sensitive data, secure neural network inference (SNNI) entails that only the client learns the output; the input remains the client's secret and the neural network's parameters remain the service provider's secret. Several SNNI approaches were proposed and evaluated in environments where both service providers and clients used powerful computers. In many real settings, for instance in edge computing, client devices are resource-constrained. This paper is the first to investigate the impact of client-side resource constraints on SNNI. We perform experiments with two state-of-the-art SNNI approaches and three neural networks. We vary the compute and memory capacity of the client device and measure the impact on inference time. Our findings show that client-side resource constraints significantly impact the performance and even the applicability of SNNI approaches. The results indicate the limits of current SNNI approaches for resource-constrained clients. Based on the results, we identify research directions to improve SNNI for resource-constrained clients.

Neural networks (NNs) have become one of the most important tools for artificial intelligence. Well-designed and trained NNs can perform inference (e.g., make decisions or predictions) on unseen inputs with high accuracy. Using NNs often involves sensitive data: Depending on the specific use case, the input to the NN and/or the internals of the NN (e.g., the weights and biases) may be sensitive. Thus, there is a need for techniques for performing NN inference securely, ensuring that sensitive data remain secret. In the past few years, several approaches have been proposed for secure neural network inference. These approaches achieve better and better results in terms of efficiency, security, accuracy, and applicability, thus making big progress toward practical secure neural network inference. The proposed approaches make use of many different techniques, such as homomorphic encryption and secure multi-party computation. The aim of this article is to give an overview of the main approaches proposed so far, their different properties, and the techniques used. In addition, remaining challenges toward large-scale deployments are identified.