In this paper we have tried to design a prototype of Farsi/English search engine. It has the duty of covering the web features such as heterogeneity, volatility and huge amount of unstructured information. These features as well as the rapid advance in technology, challenge the classical Information Retrieval (IR) techniques.

Although a growing number of Farsi-supported sites exist, still few research works have been done regarding the encoding and indexing of Farsi texts. It seems that Unicode is sufficiently capable of preparing a conclusive environment within this respect specially regarding to indexing web pages, however Many common Farsi code-pages have to be converted into Unicode, in order to cover most of the existing sites.

We utilized Unified Modeling Language (UML) to generate a visual easy-to-scale model, and to assure scalability and reliability, we used Clustering techniques and RAID. We've tried to apply Common Object Request Broker Architecture (CORBA) due to distributed object-oriented design and our agent-oriented trends in the system.

Keywords: Unicode, UML, CORBA, Information Retrieval, Search engine, Farsi language, clustering