An openflow large-scale flow table elastic energy saving and efficient search system and method

Abstract

The present invention discloses that the present invention provides an OpenFlow large-scale flow table elastic energy-saving and high-efficiency search architecture and method, including: an active and accurate flow layer for caching active and accurate flow entries in the network, and realizing high-speed and low-power flow table caching Search; the aggregation flow layer is used to store wildcard flow entries to alleviate the problem of insufficient TCAM storage capacity and improve the TCAM cache hit rate; the aggregation flow layer includes TCAM and DRAM; the individual flow layer is used to store data that does not meet the aggregation conditions temporarily Accurate flow entries to improve the packet forwarding capability of the OpenFlow switch; the individual flow layer includes SRAM and DRAM. The method of the invention uses a cross-linked list to build a sparse grid, dynamically applies for cache space, can store all accurate streams that meet the conditions, does not have empty table items, has high space utilization, can adapt to dynamic changes in network traffic, and is flexible .

Description

An OpenFlow large-scale flow table elastic energy-saving and efficient search system and method technical field The present invention relates to the field of OpenFlow flow table, be specifically related to a kind of OpenFlow large-scale flow table elastic energy saving and high Efficient search system and method. Background technique [0002] Software Defined Networking (SDN), as a new type of network architecture, The control functions in the traditional forwarding equipment are extracted and centralized in the SDN controller to manage all the data planes. There are forwarding devices to realize flexible control of network traffic. The SDN controller formulates flow rules based on the global network view, It is uniformly distributed through the southbound protocol interface, and is sent to the switch of the data plane for learning and caching to generate a local rule set. It directs the processing of network packets. The OpenFlow protocol is one of the most widely...

AI Technical Summary

Problems solved by technology

Existing energy-saving solutions often need to pre-set the size of the flow cache, which is poor in scalability and is prone to the following problems: If the energy-saving cache is set too small, the number of flow entries it can accommodate is limited, which will cause some important flow entries to be hashed. If the energy-saving cache is set too large, although it can accommodate all important flow entries, due to the unevenness of the hash algorithm, there will be many empty tables in the cache. items, resulting in a waste of cache space

[0007] 2. Large-scale flow table lookup performance problem

This prior art solution does not optimize the storage of the cache space well, so that the utilization rate of the cache space is not high, and due to insufficient cache space, table entries are frequently replaced

Images