Cloud File Caching with Priority-Based Eviction

The patent implements dynamic caching where files are cached in local memory based on real-time conditions such as network availability, storage capacity, and user activity patterns. The system continuously monitors local storage usage and automatically adjusts the cached file set, evicting files when thresholds are exceeded and caching new files when capacity allows, thereby maintaining offline availability while adapting to changing storage constraints.

The system changes the state of file caching by transitioning files between cached and non-cached states based on multiple parameters including file access frequency, user preferences, network conditions, and available storage capacity. This parameter-driven approach allows the system to optimize offline availability by caching frequently accessed files while evicting less important ones when storage is constrained.

The patent applies different caching strategies and priority levels to different files based on their specific characteristics such as file type, size, access patterns, and user importance ratings. Instead of treating all files uniformly, the system assigns local quality attributes to each file that determine its caching behavior, allowing high-priority small files to be cached aggressively while low-priority large files use different strategies, thus achieving fast access for critical files while maintaining flexibility for diverse file types.

The system segments the file cache into multiple priority levels and categories (e.g., high priority, medium priority, low priority; frequently accessed, occasionally accessed, rarely accessed). This segmentation allows the system to manage diverse file types and sources effectively by applying appropriate caching policies to each segment, maintaining fast access for important files while preserving adaptability for handling various file categories with different requirements.