The compression of data is very important especially when the amount of data is very large in order to reduce the storage size required by the original data, to give faster search and retrieval of the data, and finally to give some level of security since' the information is stored in a coded form. In this research we selected huffman codes algorithm because of its features of providing unique codes for each character and gives instantaneous decoding as soon as the codes arrives to the receiver. This algorithm is used,in this research to reversibly compress prolog programs and prolog databases with optimum results and processing measures in terms of speed, size, and accuracy. A considerable part of the research was dedicated towards the Intelligent File Store (IPS) which is a proposal for a smart database machine whose aim is both to reduce the software required for the database applications and to anticipate some of the demands for future knowledge based systems. The IPS uses the Binary Relationship Model (BRM) as its underlying means for representing information. In database terms the BRM has but two simple concepts: entities (which may be non-lexical or lexical) and binary relationship (which are two way links between two entities). Information may be modeled in terms of triples of the form: (entity, relationship, entity) The aim of this research is to design computer software model for the compression programs using huffman codes and to store the data in a triple store memory, to reduce the space needed to store data. The proposed system was implemented Turbo-pascal version 6.0 on the IBM microcomputer.