Research On Classifiable Index Of Forwarding Information In Content Routers

With the rapid development of Internet informatization,important technologies such as cloud computing,big data,and edge computing continue to appear in various fields,and Internet services are gradually turning to information content services.The lack of IP addresses and poor security are gradually exposed by TCP/IP network.Therefore,a novel network architecture named Named Data Networking is proposed.However,there are still a series of problems and challenges to be solved,especially the effective aggregation of forwarding information,the fast lookup for name,the high capacity and the support for the name matching mechanisms.Faced with these problems,research on classifiable index of Forwarding Information in Content Routers is proposed in this dissertation.An efficient index structure that supports the classification and aggregation of forwarding information is proposed,and a high-performance,applicable deployment PIT storage structure and its name lookup algorithm are designed.Some innovative works are carried out below.Firstly,an index structure C&I based on character convolutional neural network is proposed.C&I learns the distribution law of the semantic information and the distribution of data in static memory to obtain the function between data and the logical index address,which achieves more uniform mapping and more aggregated data distribution,thereby supporting the actual needs of different application scenarios.The simulation results show that C&I has excellent performance in terms of model prediction accuracy,memory consumption,and data aggregation.Secondly,an effective Pending Interest Table based on C&I,called C&I-PIT is proposed,which can perform fast name lookup for variable-length name,store forwarding information efficiently,support Interest Flooding Attack detection,and support the mechanism of All Name Prefix Matching and Exact Name Matching.The simulation results show that C&I-PIT can aggregate name with the same name prefix effectively,locate malicious prefixes quickly and accurately,and support Interest Flooding Attack detection effectively.When the amount of data reaches millions,the C&I-PIT can be deployed on SRAM to realize effective forwarding information storage.At the same time,the throughput of C&I-PIT is 1.42 million packets per second,which can quickly process millions of data packet requests in the network.