Background
Artificial intelligence is becoming one of the fastest-growing consumers of computational resources worldwide. As AI systems scale, concerns surrounding energy consumption, sustainability and environmental impact are becoming increasingly significant.
Future cities will require intelligent infrastructure capable of supporting advanced AI applications while minimising their carbon footprint.
Research Challenge
As Large Language Models (LLMs), AI agents and autonomous systems become increasingly integrated into everyday life, the computational infrastructure required to support them is growing rapidly. While cloud and edge computing platforms enable powerful AI-driven services, they also consume significant amounts of energy, creating financial and environmental challenges for future digital infrastructure.
A key challenge is determining how AI services should be deployed across distributed cloud and edge environments to balance performance, resource utilisation and sustainability objectives. Decisions that improve latency and Quality of Service (QoS) may increase energy consumption, while energy-saving strategies may negatively impact application performance. Understanding and managing these competing objectives is a major challenge for next-generation AI systems.
This project investigates how intelligent deployment and resource management techniques can reduce the environmental impact of AI services while maintaining the responsiveness required by real-time applications. The research will explore challenges such as energy-aware service placement, carbon-aware computing, adaptive workload migration, resource optimisation and multi-objective decision-making using techniques such as Reinforcement Learning.
Example application domains include smart-city digital services, autonomous transportation systems, AI-powered environmental monitoring, large-scale AI agent platforms and edge-hosted generative AI applications. The work will involve designing and evaluating optimisation strategies that enable AI systems to operate more efficiently while reducing their overall energy consumption and carbon footprint.
Topics
Sustainable AI; Green computing; Edge computing; Multi-objective optimisation; Reinforcement learning; Carbon-aware computing
Impact
People are increasingly concerned about the environmental consequences of emerging technologies. This project addresses the challenge of ensuring that the next generation of AI systems can be both powerful and sustainable, helping to create greener and more responsible digital infrastructure.