NameFilter: Achieving fast name lookup with low memory cost via applying two-stage Bloom filters

Fast Name Lookup based on Component State Transfer in Content Centric Networking

“…Wang proposed two-stage Bloom filter called NameFilter, in the first stage it determines the greatest number of components of name, it looks up the prefix in a group of Bloom filter in the second stage [7]. Prefixes are mapped into different Bloom filter depending on forwarding port, which significantly reduced the memory cost.…”

IP Address Lookup Algorithm Using a Vectored Bloom Filter

“…Hence, the IP address lookup has become one of the most challenging functionalities that need to be performed at wire-speed. Various IP address lookup algorithms have been studied such as trie-based [4][5], hashing-based [6], and Bloom filter-based algorithms [7][8][9]. Because of their sizes, the trie and the hash table are usually stored using off-chip memories, and IP address lookup procedures are completed through off-chip memory accesses.…”

“…However, an access to an off-chip memory is 10 to 20 times slower than access to an on-chip memory [10]. An on-chip Bloom filter has been used to reduce the number of off-chip memory accesses [7][8][9].…”

Steerable Name Lookup based on Classified Prefixes and Scalable One Memory Access Bloom Filter for Named Data Networking

“…The number of probe times equals to the number of component to be contained in the prefix at the worst case to make difficult cope with data packet. Wang et .al proposed two-stage Bloom filter, in the first stage it determines the greatest number of components of name and in the second stage it looks up the prefix in a group of Bloom filter [6]. Prefixes are mapped into different Bloom filter depending on forwarding port, which significantly reduced the memory cost.…”