Dalarna University's logo and link to the university's website

The users of cloud services prioritize cost and performance, but they increasingly demand sustainable practices. Sustainability is no longer a choice for businesses but a strategic imperative that shapes global industries. This paper presents a new system for utilizing the render farms in cloud data centers. The system aims to reduce energy consumption and costs in cloud data centers while maintaining a specific level of performance, particularly when rendering images and videos. The system can be described as a cloud-based expert system that offers rendering as a service, while considering user preferences for performance, cost, and energy efficiency. The system reads different scene rendering parameters and accordingly chooses the most suitable GPUs that fit the user's requirements. In other words, the system inputs are the scene complexity and user preferences. The output is the optimal GPU for rendering. Scene complexity is determined based on several parameters, such as the number of frames and polygons, resulting in one scene-related value. The user preferences are also normalized to a preferences-related value. Then, these values are aggregated to determine the optimal available GPU to render the scene at the lowest cost, minimum possible energy consumption, and highest performance. © 2024 IEEE.