Big Data Visualization and Processing For Multiuser Virtual Reality Educational Applications

Virtual Reality (VR) educational applications struggle with massive amounts of data, which pose significant challenges for real-time processing and analysis (e.g., in domains such as biology education, where complex network visualization is required). This study presents a comprehensive, large-scale data analytics framework for multiuser VR educational applications, demonstrated by an innovative gene interaction network model representing 100 diseases and 400 genes from the NCBI database. The framework addresses critical large-scale data challenges, including volume (massive interaction datasets), velocity (real-time processing), variety (heterogeneous data types), and veracity (data quality assurance), in immersive learning environments. Advanced techniques, including distributed computing, stream processing, and optimized data structures, were implemented alongside performance optimization strategies, including asynchronous compute buffer reading, burst compiling, and single-pass stereo rendering. Evaluation with 20 participants demonstrated exceptionally large-scale data-handling capabilities: a 99.29% task success rate while processing 2.5 GB of interaction data per hour; an 86.375 System Usability Scale score with real-time analytics; 98.2% interactive accuracy across seven VR actions with continuous data streaming; and 90 FPS performance with 8.9ms latency despite intensive data processing.