{"id":9,"date":"2025-09-04T19:43:24","date_gmt":"2025-09-04T18:43:24","guid":{"rendered":"https:\/\/soc2.ucd.ie\/?page_id=9"},"modified":"2025-09-05T00:48:19","modified_gmt":"2025-09-04T23:48:19","slug":"about-soc2-lab","status":"publish","type":"page","link":"https:\/\/soc2.ucd.ie\/index.php\/about-soc2-lab\/","title":{"rendered":"About SOC\u00b2 Lab"},"content":{"rendered":"\n

Digital services and AI models now power everything from city traffic control to personalised healthcare, but their energy use and carbon footprint are growing fast. SOC\u00b2 Lab develops sustainable ways to orchestrate these services across the edge\u2013to\u2013cloud continuum<\/strong>, enabling low-carbon, resilient, and intelligent systems for smart cities, smart buildings, and global-scale applications<\/strong>.<\/p>\n\n\n\n

Our work combines energy-efficient, carbon-aware computing<\/strong> with cutting-edge paradigms such as serverless architectures, edge AI, and distributed large language models. From real-time urban analytics to building energy optimisation, we design orchestration models and algorithms that reduce environmental impact while improving performance and resilience, directly contributing to UN Sustainable Development Goals 7, 9, 11, and 13.<\/strong><\/p>\n\n\n\n

\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n<\/figure>\n\n\n\n

Research Focus<\/h2>\n\n\n\n
    \n
  1. Sustainable Orchestration in Edge\u2013Cloud and Multi-Cloud Environments<\/strong>: We design mechanisms for energy-efficient, carbon-aware orchestration across diverse computing infrastructures, from edge devices to federated and multi-cloud platforms. Our work covers dynamic service placement, task scheduling, and resource provisioning that balance performance, cost, and environmental impact, leveraging advanced networking platforms such as MEC, B5G, 6G, and SAGIN to enable next-generation resilient and sustainable urban infrastructures.<\/li>\n\n\n\n
  2. LLM@Edge and Distributed LLMs<\/strong>: We investigate the deployment and optimisation of large language models at the network edge to deliver low-latency, resource-efficient AI services. Our research also advances distributed LLM architectures, including Mixture of Experts (MoE) models, enabling scalable and sustainable AI across the edge\u2013cloud continuum. Applications include real-time urban analytics, building energy optimisation, and other sustainability-driven domains.<\/li>\n<\/ol>\n\n\n\n

    Our Values<\/h2>\n\n\n\n