Task Partitioning and Scheduling for Fractal Graphics Rendering Based on Web Worker

As web technologies shift from traditional content delivery toward high-performance computing, Web Worker-based parallelism for compute-intensive tasks such as fractal rendering has become essential for improving web application performance and interactivity. However, due to memory isolation and limited hardware access imposed by the browser sandbox, existing parallel approaches often struggle with excessive communication overhead, severe load imbalance, and straggler blocking when rendering spatially heterogeneous fractal graphics, especially under high-frequency interaction. This paper presents a task partitioning and scheduling scheme for parallel fractal rendering on Web Worker that addresses these challenges. By introducing load prediction for fractal rendering tasks, a communication-aware task partitioning scheme is achieved together with a scheduling strategy that combines interaction-aware adaptive prioritization and stale task cancellation. Experimental results show that the proposed method effectively reduces task skew and communication costs while outperforming baseline methods in term of parallel performance, load balancing, and response latency. Under high-frequency interaction, it also delivers notable improvements in system responsiveness and interaction smoothness, substantially enhancing both rendering efficiency and user experience for complex fractal graphics on the web.