Search on the web is a daily activity for many people through- out the world. Applications based on search are everywhere. We are in a data rich situation this becomes an obstacle for Information retrieval system. In this paper, we propose a pipelining architecture of indexing in order to enhance memory utilization and fast query optimization and also a soft clustering algorithm is applied to group documents into clusters hierarchically linked. Each cluster is labeled with is most relevant term known as document identifier. Such that documents within the same cluster are similar. In this way it will create hierarchy of index so that search will travel from lower level to higher level. The comparisons of the document against the user query will direct to specific clusters not the whole collection of clusters.

Agglomerative Clustering Pipelining Hierarchical Indexing