China
Africa
Explore chapters and articles related to this topic
Mapping Network Device Functions to the OSI Reference Model
Published in James Aweya, Designing Switch/Routers, 2023
The adoption of Classless Inter-Domain Routing (CIDR) [RFC4632] means that a routing lookup needs to perform a “longest prefix match” search in the forwarding table. A router maintains a set of IP destination address prefixes (of variable length) in a forwarding table, each associated with a next-hop IP address and outbound interface (plus, possibly, other important information as discussed in Chapters 2 and 5). Given a packet, the “longest prefix match” search involves finding the longest prefix in the forwarding table that matches the first contiguous set of bits of the IP destination address of the packet.
Routing and Addressing
Published in Rui Valadas, OSPF and IS-IS, 2019
When a packet arrives at a router, the router performs a forwarding table lookup and tries to match the destination IP address contained in the packet header with one of its forwarding table entries. The matching verification for an incoming IP address is performed at each forwarding table entry by comparing (i) the entry prefix with (ii) the prefix obtained by applying the entry subnet mask to the incoming IP address. There can be more than one forwarding table entry matching a given IP address. When this happens, the entry with the longest prefix is selected for routing; this is known as the longest prefix match rule.
TCAM-Based Forwarding Engine
Published in Weidong Wu, Packet Forwarding Technologies, 2007
A mask bit string is used to indicate the portion of a stored data string in parallel comparison and the length of prefix (called horizontal AND operation). VLMP is logical OR operation applying to corresponding bits from different mask strings to obtain the maximum length of prefixes matching a given key (called vertical OR operation). Once the result of VLMP is obtained, the longest prefix match entry is found.
An efficient and DoS-resilient name lookup for NDN interest forwarding
Published in Connection Science, 2021
Dacheng He, Dafang Zhang, Yanbiao Li, Wei Liang, Meng-Yen Hsieh
Efficient data transmission and reliable security models are essential for interest forwarding in NDN. Researchers have proposed several throttle mechanisms to limit malicious stream. However, they reduce standard stream transmission and affect the normal forwarding process in this way. Unlike that mechanism, accelerating data forwarding can also relieve damages caused by DoS without affecting standard stream transmission. As name lookup is the core operation of NDN forwarding, a fast and efficient name lookup algorithm can improve data transmission and resist DoS attacks. Since NDN makes content centric in data transmission, and the contents’ structure is far more complicated than the design structure of IP addresses, it is still a big challenge to increase NDN forwarding efficiency. Most existing methods store name prefixes (just like IP prefixes) into different hash tables based on their length (short for component number); they generally conduct name lookup processes just like longest prefix match (LPM) (Huang et al., Apr 10-15, 2012; Li et al., Sep 2-4, 2014) in TCP/IP network, called the longest name prefix match (LNPM). During this process, most of the costs are incurred on hash lookup and string match processes, and therefore it is a crucial factor to reduce the consumption in the NDN forwarding process.