Flow table search method and device

Abstract

The embodiment of the invention provides a flow table search method and device. The method comprises the following steps: a five-component set of a received message is obtained, wherein the five-component set comprises a source IP (internet protocol) address, a destination IP address, an IP protocol type, a source port number and a destination port number; after the five-component set of the message is arrayed sequentially according to the size of the source IP address and the size of the destination IP address, the sequence of the re-arrayed five-component set of the message is used as a keyword for searching the flow table, so as to judge whether a certain table entry in the flow table is hit; and if a certain table entry in the flow table is hit, the transmission direction of the message is determined according to the size of the source IP address and the size of the destination IP address, and the content of a session table is searched. In the embodiment of the invention, the forward and reverse flow tables can be searched by using the same keyword, so that the forward and reverse flow tables can be stored by one flow table entry; the content space can be saved, the utilization rate for searching the memory is improved, and the method can be used in an NP (network processor), a CPU (central processing unit), an FPGA (field programmable gate array) and an ASIC (application specific integrated circuit).

Description

technical field [0001] The present invention relates to the field of network technology, in particular to a flow table search method and device. Background technique [0002] With the development of the network, the network bandwidth is increasing day by day, and the use of a general-purpose central processing unit CPU (Central Processing Unit) cannot meet the demand for network bandwidth. , Network Processer), application specific integrated circuit (ASIC, Application Specific Integrated Circuit), field programmable gate array (FPGA, Field Programmable Gate Array), etc., in data forwarding, the general and stable parts are processed by special processing devices, complex and The volatile part of the demand is handed over to the general-purpose CPU for processing. [0003] The current service forwarding models of network devices can be roughly divided into two types, one is packet-based forwarding, and the other is flow-based forwarding. [0004] Packet-based forwarding is...

AI Technical Summary

Problems solved by technology

Existing technology 2 has a session table. When the forward flow table is established, the reverse flow table is established at the same time, and the session table is established at the same time. The bidirectional flow table needs to have the index of the session table, and the index of the bidirectional flow table is stored in the session table. The session table Two flow tables are required, and association is difficult. Aging will only delete flow tables in one direction. If you need to delete flow tables in two directions, the processing is complicated. You need to take out the session table index from the flow table in the flow table and then take out the flow in the other direction. table index to delete the flow table in the other direction

[0009] In the process of realizing the present invention, the inventor found that there are at least the following problems in the prior art: it is necessary to establish two forward and reverse flow tables, and a session table is used to associate the two entries, and two flows need to be reserved in the session table The information of the table is used to delete the forward and reverse flow tables. Since the number of flow tables is very large, the forward and reverse flow tables retain a lot of redundant information and consume a lot of memory space

Images