An information retrieval system attempts to find relevant documents to respond to the user request. An internet search engine (e.g. Google) returns thousands of matched documents, some are relevant and others irrelevant to a text query. Text has become the most important data, because large-scale (heterogeneous) electronic text collections are more available now ever than before.  As a result, text data retrieval still requires a tremendous amount of processing time because of the size of the data and the complexity of information retrieval algorithms. The solution to this problem may come from hardware supported information retrieval algorithms.

There were number of hardware acceleration techniques proposed for databases implementation. Text retrieval is the most know.  But most of the hardware supported database accelerators became absolute (But none of these approaches gain significant attention) due to continues changes and improvement in information retrieval algorithms. However, Reconfigurable Computing may adopt frequent hardware modifications through its tailorable hardware. In ASIC, acceleration is achieved through fixed number of the parallel VLSI units. On the other hand RCs exploits parallelism for a given application through reconfigurable and flexible hardware units. The degree of the parallelism can be tuned for data.