Ecoscape as a Validated Edge-Cloud-Continuum Simulator: An LLM Use-Case Integration and Comparative Study
The rapid proliferation of Edge Computing demands robust and accurate simulation tools to model and optimize the complex interactions across the Edge-Cloud-Continuum. This thesis addresses the pressing need for a structured evaluation of the current simulation landscape by presenting a comparative study of ten representative state-of-the-art edge computing tools, leading to the deduction of a comprehensive requirements catalogue. This guide establishes a future-proof benchmark for evaluating and developing next-generation edge simulation tools. Using this newly defined standard, the thesis conducts a targeted validation of Ecoscape. The analysis revealed that Ecoscape satisfies nearly all requirements, demonstrating its high technical maturity, but lacked sufficient representative use-cases to fully showcase its adaptability to emerging technologies. To address this gap and further validate its utility, this work introduces a novel, timely use-case: A Large Language Model (LLM) workload deployment across the continuum. The integration and subsequent experimental evaluation of this LLM scenario successfully demonstrate Ecoscape’s capability to model complex, resource-intensive and modern computing paradigms.
Preview
Rights
Use and reproduction:
Please note that individual components of the publication may be subject to other licensing or copyright conditions.
