Towards Low-Latency Service Delivery in a Continuum of Virtual Resources: State-of-the-Art and Research Directions

Abstract

The advent of softwarized networks has enabled the deployment of chains of virtual network and service components on computational resources from the cloud up to the edge, creating a continuum of virtual resources. The next generation of low latency applications (e.g., Virtual Reality (VR), autonomous cars) adds even more stringent requirements to the infrastructure, calling for considerable advancements towards cloud-native micro-service-based architectures. This article presents a comprehensive survey on ongoing research aiming to effectively support low latency services throughout their execution lifetime in next-generation networks. The current state-of-the-art is critically reviewed to identify the most promising trends that will strongly impact the full applicability and high performance of low latency services. This article proposes a taxonomy as well as specific evaluation criteria to classify research across different domains addressing low latency service delivery. Current architectural paradigms such as Multi-access Edge Computing (MEC) and Fog Computing (FC) alongside novel trends on communication networks are discussed. Among these, the integration of Machine Learning (ML) and Artificial intelligence (AI) is introduced as a key research field in current literature towards autonomous network management. A discussion on open challenges and future research directions on low-latency service delivery leads to the conclusion, offering lessons learned and prospects on emerging use cases such as Extended Reality (XR), in which novel trends will play a major role.