Packet classification is to classify incoming packets into "class" in the Internet routers in order to provide different services according to classes that packet belongs. Packet classification is one of the most challenging operations since it should be performed in wire-speed referring the thousands of pre-defined multi-field rules. In this paper, we proposed a new packet classification architecture. The proposed scheme is based on parallel multiple hashing for source-destination prefix tuples and linked-list search for remaining fields on rules with the same source and destination prefixes. Prefix grouping is also proposed to reduce the number of tuples. Simulation results using real databases show that the proposed scheme requires small number of memory accesses and O(N) memory space for N rule classifier. The proposed scheme also has very good characteristics in rule updating and scalability toward large classifier.