Descrição:
This dataset captures detailed context metrics from users, servers, and network infrastructure during adaptive video streaming sessions in a simulated Edge-Cloud environment. Collected as part of a learning-based architecture prototype, it supports research in Quality of Experience (QoE) optimization, resource orchestration, and adaptive delivery strategies. The dataset includes user-side playback metrics, server resource usage, and network conditions, reflecting realistic scenarios with dynamic load and user movement. Built upon HTTP Adaptive Streaming (HAS) using dash.js and enhanced by Content Steering Services (CSS) and machine learning-driven orchestration, the system enables dynamic container deployment at the edge to prevent SLO violations and sustain QoE. This dataset is obtained from the doctoral thesis for evaluating video streaming systems under varying network and service conditions.