X4-MATCH: Sustainable Prediction-based Distribution of Video Encoding on Cloud and Edge
Samira Afzal; Narges Mehran; Andrew C Freeman; Manuel Hoi; Armin Lachini; Christian Timmerer; Radu Prodan (2026): X4-MATCH: Sustainable Prediction-based Distribution of Video Encoding on Cloud and Edge In: 2026 IEEE International Parallel and Distributed Processing Symposium (IPDPS).
The rapid expansion of video traffic has made it one of the most energy-intensive workloads on cloud and edge infrastructures. As encoding remains essential for streaming, gaming, and immersive applications, efficient task scheduling is required to balance service quality, cost efficiency, and sustainability. In this work, we present X4-MATCH, a sustainable scheduling framework that integrates machine learning–based prediction with game-theoretic heuristics to optimize video encoding workloads across cloud–edge infrastructures. The framework formulates four performance optimization objectives to balance performance and sustainability goals in video encoding: 1) processing and transmission time, 2) price, 3) energy use, and 4) CO2 emissions. X4-MATCH leverages the Extra-Trees regressor model to predict video encoding performance metrics, integrated with a matching theory strategy for mapping media provider encoding workloads onto resource provider computing resources. We experimentally validate X4-MATCH on a real-world testbed incorporating Amazon Web Services cloud virtual machine instances and local edge servers. Results show that X4-MATCH outperforms state-of-the-art methods by reducing video encoding time by 63.3%, price by 54.2%, energy by 26.8%, and CO2 emissions by 10.3% compared to state-ofthe-art methods. The X4-MATCH implementation is publicly available at: https://doi.org/10.5281/zenodo.18525889.
