CA-LAP: A Scheduling Model for Video Streaming Data Transcoding Tasks in a Distributed Environment

For the task scheduling optimization of video stream transcoding in distributed environment, the traditional scheduling strategy is difficult to achieve load balancing. In this paper, we propose a distributed video streaming transcoding CA-LAP load balancing strategy, which is based on Dubbo and Zookeeper architecture, and uses the consistent hashing method to map transcoding requests to a hash ring of node hashes to initially achieve load balancing. Based on this, the least active priority method is applied and improved to achieve dynamic scheduling of local nodes around the request, and the availability of the selected nodes is guaranteed using the DVR policy. The experimental results show that the proposed CA-LAP policy shows superiority in node load variance metrics for large-scale distributed video stream transcoding compared to using only the consistent hashing method, and achieves better application results in real transcoding scenarios. This study provides a new load balancing solution for the field of distributed video stream transcoding, which helps to improve system performance and scalability.