With the increasing growth of online video streaming platforms, it has become imperative to ensure that viewers enjoy a seamless and exceptional viewing experience. Adaptive bitrate streaming is a crucial technology for optimizing video transmission by continuously adjusting video quality based on network conditions and device capabilities. However, there are significant challenges in maintaining a consistent Quality of Experience (QoE) over a distributed network of Adaptive Bitrate (ABR) video servers, particularly when there are varying user demands and network congestion. This research proposes an architectural design that can enhance users’ QoE by load balancing for ABR video streaming servers and efficiently distributing incoming video requests among scalable service components by predictive analysis, and real-time network monitoring. The design aims to minimize issues that degrade the QoE, such as buffering and fluctuations in video quality. This can be achieved by employing dynamic resource allocation, considering variables such as user demands and server workload. Dynamic load balancing policies in accordance with different scenarios will consequently improve the viewers’ experience, increase satisfaction and thereby enhance the overall streaming performance. To evaluate the performance and efficiency of the proposed workload-aware architecture, comprehensive simulations for diverse streaming scenarios are created, encompassing both synthetic and real-world ABR video datasets.