Large file operations support using order preserving perfect hashing functions

Most computer applications require efficient management of data and fast execution of basic file operations over large data volumes. Specifically, in real-time environment applications are faced with severe constraints for total execution time of basic operations. This paper introduces a method for physical organization of large database files, based on an order preserving hashing scheme. The hashing scheme combines two functions: an order preserving and an ordinary hashing function. An original set of algorithms take advantage of implemented physical organization to achieve efficient basic file operations. The proposed method guarantees retrieval of any record in a single disk access, and minimum number of disk accesses for range search and key sequential operations for large dynamic files.