Prefix Matching Research Articles

XOR-Based Schemes for Fast Parallel IP Lookups

An IP router must forward packets at gigabit speed in order to guarantee a good quality of service. Two important factors make this task a challenging problem: (i) for each packet, the longest matching prefix in the forwarding table must be quickly computed; (ii) the routing tables contain several thousands of entries and their size grows significantly every year. Because of this, parallel routers have been developed which use several processors to forward packets. In this work we present a novel algorithmic technique which, for the first time, exploits the parallelism of the router also to reduce the size of the routing table. Our method is scalable and requires only minimal additional hardware. Indeed, we prove that any IP routing table T can be split into two subtables T1 and T2 such that: (a) |T1| can be any positive integer k ≤ |T| and |T2| ≤ |T| - k + 1; (b) the two routing tables can be used separately by two processors so that the IP lookup on T is obtained by simply XOR-ing the IP lookup on the two tables. Our method is independent of the data structure used to implement the lookup search and it allows for a better use of the processors L2 cache. For real routers routing tables, we also show how to achieve simultaneously: (a) |T1| is roughly 7% of the original table T; (b) the lookup on table T2 does not require the bestmatching prefix computation.

Efficient Multimatch Packet Classification and Lookup with TCAM

Today's packet classification systems are designed to provide the highest-priority matching result, such as the longest prefix match, even if a packet matches multiple classification rules. However, new network applications demanding multimatch classification - that is, requiring all matching results instead of only the highest-priority match - are emerging. Ternary content-addressable memory is becoming a common extension to network processors, and its capability and speed make it attractive for high-speed networks. The proposed TCAM-based scheme produces multimatch classification results with about 10 times fewer memory lookups than a pure software approach. In addition, their scheme for removing negation in rule sets saves up to 95 percent of the TCAM space used by a straightforward implementation.

Simple and fast IP lookups using binomial spanning trees

High performance Internet routers require an efficient IP lookup algorithm to forward millions of packets per second. Various binary trie data structures are normally used in software-based routers. Binary trie based lookup algorithms are simple not only for searching for the longest prefix match but also for updating the routing table. In this paper, we propose a new IP lookup algorithm based on binomial spanning trees. The proposed algorithm has the same advantages of simple search and update processes as the ones based on binary trie. However, the performance of the proposed algorithm in terms of memory requirement and the lookup time is better than the schemes based on binary tries, such as path-compression and level-compression.

Efficient IP forwarding engine with incremental update

Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP...

고속 라우터를 위한 향상된 비트맵 룩업 알고리즘

인터넷 회선의 고속화에 따라 패킷을 초당 기가비트 이상의 속도로 전송하는 라우터에 대한 연구가 활발하다. 본 논문에서는 라우터의 병목인 최장 프리픽스 검색(LPM : longest prefix matching)을 효율적으로 수행하기 위해 제안된 비트맵 트라이(Bitmap Trie) 알고리즘을 혁신적으로 향상시킨 방법을 제시한다. 이 방법은 검색시간을 단축하기 위하여 다음과 같은 세 가지 기법을 적용하였다. 첫째, 카운트 테이블을 추가하여 기존 알고리즘에서의 과도한 시프트 연산을 제거하여 프로세서 내의 처리시간을 단축했다. 두 번째는 트랜스퍼 테이블내의 중복된 포워딩 정보를 제거하여 메모리 효율을 증가시켰으며, 마지막으로, 포워딩 정보에 대한 접근빈도에 따라 검색범위를 다원화하여 데이터 접근을 최적화하였다. 또한, 신뢰도가 가장 놀은 실행-구동 시뮬레이션 방법을 채택함으로써 검색의 지연요소를 발생 원인별로 세분하여 알고리즘이 실행되는 과정을 면밀하게 분석할 수 있었다. 실험 결과는 실제 시스템으로부터 실측한 결과와 비교하는 검증과정을 거쳤으며, 그 결과 기존 알고리즘에 비해 검색 수행시간을 약 82% 단축한 알고리즘을 개발하였다. As the Internet gets faster, the demand for high-speed routers that are capable of forwarding more than giga bits of data per second keeps increasing. In the previous research, Bitmap Trie algorithm was developed to rapidly execute LPM(longest prefix matching) process which is Well known as the Severe performance bottleneck. In this paper, we introduce a novel algorithm that drastically enhanced the performance of Bitmap. Trie algorithm by applying three techniques. First, a new table called the Count Table was devised. Owing to this table, we successfully eliminated shift operations that was the main cause of performance degradation in Bitmap Trie algorithm. Second, memory utilization was improved by removing redundant forwarding information from the Transfer Table. Lastly. the range of prefix lookup was diversified to optimize data accesses. On the other hand, the processing delays were classified into three categories according to their causes. They were, then, measured through the execution-driven simulation that provides the higher quality of the results than any other simulation techniques. We tried to assure the reliability of the experimental results by comparing with those that collected from the real system. Finally the Enhanced Bitmap Trie algorithm reduced 82% of time spent in previous algorithm.

An O(log n) dynamic router-table design

Internet (IP) packet forwarding is typically done by finding the longest prefix in a router table that matches the packet's destination address. For W-bit destination addresses, the use of binary tries enables us to determine the longest matching prefix in O(W) time, independent of the number n of prefixes in the router table. New prefixes may be inserted and old ones deleted in O(W) time also. Since n/spl Lt/2/sup W/ in real router tables, it is desirable to develop a data structure that permits longest prefix matching as well as the insertion and deletion of prefixes in O(logn). These three operations can be done with O(logn) cache misses using a B-tree data structure [S. Suri et al., (2001)]. However, the runtime (including operation cost and cost of cache misses) is not O(logn). In this paper, we develop a data structure in which prefix matching, prefix insertion, and deletion can each be done in O(logn) time. Experiments using real IPv4 routing databases indicate that, although the proposed data structure is slower than optimized variable-stride tries for longest prefix matching, the proposed data structure is considerably faster for the insert and delete operations.

Fast prefix matching of bounded strings

Longest Prefix Matching (LPM) is the problem of finding which string from a given set is the longest prefix of another, given string. LPM is a core problem in many applications, including IP routing, network data clustering, and telephone network management. These applications typically require very fast matching of bounded strings, i.e., strings that are short and based on small alphabets. We note a simple correspondence between bounded strings and natural numbers that maps prefixes to nested intervals so that computing the longest prefix matching a string is equivalent to finding the shortest interval containing its corresponding integer value. We then present retries , a fast and compact data structure for LPM on general alphabets. Performance results show that retries often outperform previously published data structures for IP look-up. By extending LPM to general alphabets, retries admit new applications that could not exploit prior LPM solutions designed for IP look-ups.

Multiway range trees: scalable IP lookup with fast updates

Internet routers forward packets based on the destination address of a packet. A packet’s address is matched against the destination prefixes stored in the router’s forwarding table, and the packet is sent to the output interface determined by the longest matching prefix. While some existing schemes work well for IPv4 addresses, we believe that none of the current schemes scales well to IPv6, especially when fast updates are required. As the Internet evolves into a global communication medium, requiring multiple addresses per user, the switch to longer addresses (e.g. IPv6) seems inevitable despite temporary measures such as network address translation boxes. Since IPv6 uses 128 bit addresses, schemes whose lookup time grows with address length (such as patricia or multibit tries) become less attractive. Because of backbone protocol instabilities, it is also important that lookup schemes be able to accommodate fast updates. In this paper, we introduce a new IP lookup scheme with worst-case search and update time of O(log n), where n is the number of prefixes in the forwarding table. Our scheme is based on a new data structure, a multiway range tree, which achieves the optimal lookup time of binary search, but can also be updated in logarithmic time when a prefix is added or deleted; by contrast, plain binary search relies on precomputation, and a single update can require O( n) time. Our performance analysis shows that, even for IPv4, multiway range trees are competitive with the best lookup schemes currently known. In fact, among existing schemes, only multibit tries have update performance comparable to our scheme and such schemes have patent restrictions. Further, when considering IPv6 or any future routing protocol that uses longer addresses, our scheme outperforms all existing schemes, including multibit tries.

High speed IP address lookup architecture using hashing

One of the most important design issues for IP routers responsible for datagram forwarding in computer networks is the route-lookup mechanism. In this letter, we explore a practical IP address lookup scheme which converts the longest prefix matching problem into the exact matching problem. In the proposed architecture, the forwarding table is composed of multiple SRAM, and each SRAM represents an address lookup table in a single prefix. Hashing functions are applied to each address lookup table in order to find out matching entries in parallel, and the entry matched with the longest prefix among them is selected. Simulation using data from the MAE-WEST router shows that a large routing table with 37000 entries is compacted to a forwarding table of 189 kbytes in the proposed scheme and achieves one route lookup every two memory accesses in average.

Reconfigurable memory architecture for scalable IP forwarding engines

In this paper, a reconfigurable memory architecture and lookup technique for IP packet forwarding engines is presented. This is achieved by partitioning the forwarding table into smaller partial lookup tables for each output port and allowing a forwarding engine to process them in parallel. This effectively reduces the complexity of finding the ‘longest prefix match’ problem to the ‘first prefix match’ problem. Our method is a flexible technique that significantly elevates the scalability of the next generation network processors and other packet processing devices. Such scalability facilitates migration to IPv6 and benefits network equipments especially in terms of growing routing table size, traffic, frequency of route updates and bandwidth requirement.

Efficient hardware architecture for fast IP address lookup

A multi-gigabit internet protocol (IP) router may receive several million packets per second from each input link. For each packet, the router needs to find the longest matching prefix in the forwarding table in order to determine the packet's next-hop. An efficient hardware solution for the IP address lookup problem is presented. The problem is modelled as a searching problem on a binary-trie. The binary-trie is partitioned into fixed size non-overlapping subtrees. Each subtree is represented using a bit-vector and can be searched in parallel for the best matching prefix in a few nanoseconds. The address lookup is implemented using a hardware pipeline with a throughput of one lookup per memory access. A distinguishing feature of the design is that forwarding table entries are not replicated in the data structure. Hence, table updates can be done in constant time with only a few memory accesses. The approach can be extended to IPv6. By applying path compression, the amount of memory required is upper bounded by O(N) where N is the number of prefixes in the routing table.

Parallelisation of trie-based longest prefix matching for fast IP address lookups

IP address lookup is an important design issue for the high performance packet forwarding engine. The forwarding engine performs a longest prefix matching on the address lookup for an incoming IP packet to determine the next hop. A parallelisation scheme of longest prefix matching algorithms based on path-compressed trie is proposed. The proposed scheme reduces memory access time of IP address lookup while keeping memory spaces of parallel elements balanced.

High-performance IP routing table lookup

Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packets forwarding, which complicates the IP lookup. Currently, this process is mainly done in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the new generation routers, in this article we propose a fast IP address lookup scheme, which significantly reduces the forwarding table to fit into SRAM with very low cost. It can also be implemented in hardware using the pipeline technique. By using our proposed method, the required memory space can be 90–170 Kb less than the previous scheme and three memory accesses for a lookup in the worst case. When implemented in hardware pipeline architecture, our mechanism can achieve one routing lookup per memory access.

An efficient IP routing lookup by using routing interval

Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packets forwarding, this complicates the IP lookup. Currently, this process is mainly performed in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the new generation routers, in this article we investigate the properties of the routing table and present a new approach for IP lookups, our approach is not based on BMP and significantly reduces the complexity, thus the computation cost of existing schemes can be significantly reduced. We also propose an efficient IP lookup algorithm, with which we improve the binary search on prefixes to 30 millions of packets per second (MPPS) and 5,000 route updates/s under the same experiment setup with an even larger routing table.

Performance modeling for fast IP lookups

In this paper, we examine algorithms and data structures for the longest prefix match operation required for routing IP packets. Previous work, aimed at hardware implementations, has focused on quantifying worst case lookup time and memory usage. With the advent of fast programmable platforms, whether network processor or PC-based, metrics which look instead at average case behavior and memory cache performance become more important. To address this, we consider a family of data structures capturing the important techniques used in known fast IP lookup schemes. For these data structures, we construct a model which, given an input trace, estimates cache miss rates and predicts average case lookup performance. This model is validated using traces with varying characteristics. Using the model, we then choose the best data structure from this family for particular hardware platforms and input traces; we find that the optimal data structure differs in different settings. The model can also be used to select the appropriate hardware configurations for future lookup engines. The lookup performance of the selected data structures is competitive with the fastest available software implementations.

Routing with a clue

We suggest a new simple forwarding technique to speed up IP destination address lookup. The technique is a natural extension of IP, requires 5 bits in the IP header (IPv4, 7 in IPv6), and performs IP lookup nearly as fast as IP/Tag switching but with a smaller memory requirement and a much simpler protocol. The basic idea is that each router adds a to each packet, telling its downstream router where it ended the IP lookup. Since the forwarding tables of neighboring routers are similar, the clue either directly determines the best prefix match for the downstream router, or provides the downstream router with a good point to start its IP lookup. The new scheme thus prevents repeated computations and distributes the lookup process across the routers along the packet path. Each router starts the lookup computation at the point its upstream neighbor has finished. Furthermore, the new scheme is easily assimilated into heterogeneous IP networks, does not require routers coordination, and requires no setup time. Even a flow of one packet enjoys the benefits of the scheme without any additional overhead. The speedup we achieve is about 10 times faster than current standard techniques. In a sense, this paper shows that the current routers employed in the Internet are clue-less; namely, it is possible to speed up the IP lookup by an order of magnitude without any major changes to the existing protocols.

Survey and taxonomy of IP address lookup algorithms

Due to the rapid growth of traffic in the Internet, backbone links of several gigabits per second are commonly deployed. To handle gigabit-per-second traffic rates, the backbone routers must be able to forward millions of packets per second on each of their ports. Fast IP address lookup in the routers, which uses the packet's destination address to determine for each packet the next hop, is therefore crucial to achieve the packet forwarding rates required. IP address lookup is difficult because it requires a longest matching prefix search. In the last couple of years, various algorithms for high-performance IP address lookup have been proposed. We present a survey of state-of-the-art IP address lookup algorithms and compare their performance in terms of lookup speed, scalability, and update overhead.

On effective multi-dimensional indexing for strings

As databases have expanded in scope from storing purely business data to include XML documents, product catalogs, e-mail messages, and directory data, it has become increasingly important to search databases based on wild-card string matching: prefix matching, for example, is more common (and useful) than exact matching, for such data. In many cases, matches need to be on multiple attributes/dimensions, with correlations between the dimensions. Traditional multi-dimensional index structures, designed with (fixed length) numeric data in mind, are not suitable for matching unbounded length string data. In this paper, we describe a general technique for adapting a multi-dimensional index structure for wild-card indexing of unbounded length string data. The key ideas are (a) a carefully developed mapping function from strings to rational numbers, (b) representing an unbounded length string in an index leaf page by a fixed length offset to an external key, and (c) storing multiple elided tries, one per dimension, in an index page to prune search during traversal of index pages. These basic ideas affect all index algorithms. In this paper, we present efficient algorithms for different types of string matching. While our technique is applicable to a wide range of multi-dimensional index structures, we instantiate our generic techniques by adapting the 2-dimensional R-tree to string data. We demonstrate the space effectiveness and time benefits of using the string R-tree both analytically and experimentally.

A Prefix Code Matching Parallel Load-Balancing Method for Solution-Adaptive Unstructured Finite Element Graphs on Distributed Memory Multicomputers

In this paper, we propose a prefix code matching parallel load-balancing method (PCMPLB) to efficiently deal with the load imbalance of solution-adaptive finite element application programs on distributed memory multicomputers. The main idea of the PCMPLB method is first to construct a prefix code tree for processors. Based on the prefix code tree, a schedule for performing load transfer among processors can be determined by concurrently and recursively dividing the tree into two subtrees and finding a maximum matching for processors in the two subtrees until the leaves of the prefix code tree are reached. We have implemented the PCMPLB method on an SP2 parallel machine and compared its performance with two load-balancing methods, the directed diffusion method and the multilevel diffusion method, and five mapping methods, the AE/ORB method, the AE/MC method, the MLkP method, the PARTY library method, and the JOSTLE-MS method. An unstructured finite element graph Truss was used as a test sample. During the execution, Truss was refined five times. Three criteria, the execution time of mapping/load-balancing methods, the execution time of an application program under different mapping/load-balancing methods, and the speedups achieved by mapping/load-balancing methods for an application program, are used for the performance evaluation. The experimental results show that (1) if a mapping method is used for the initial partitioning and this mapping method or a load-balancing method is used in each refinement, the execution time of an application program under a load-balancing method is less than that of the mapping method. (2) The execution time of an application program under the PCMPLB method is less than that of the directed diffusion method and the multilevel diffusion method.

Routing with a clue

We suggest a new simple forwarding technique to speed-up IP destination address lookup. The technique is a natural extension of IP, requires 5 bits in the IP header (IPv4, 7 in IPv6) and performs IP lookup nearly as fast as IP/Tag-switching but with a smaller memory requirement and a much simpler protocol. The basic idea is that each router adds a "clue" to each packet, telling its downstream router where it ended the IP lookup. Since the forwarding tables of neighboring routers are similar, the clue either directly determines the best prefix match for the downstream router, or provides the downstream router with a good point to start its IP lookup. The new scheme thus prevents repeated computations and distributes the lookup process across the routers along the packet path. Each router starts the lookup computation at the point its up-stream neighbor has finished. Furthermore, the new scheme is easily assimilated into heterogeneous IP networks, does not require routers coordination, and requires no setup time. Even a flow of one packet enjoys the benefits of the scheme without any additional overhead. The speedup we achieve is about 10 times faster than current standard techniques. In a sense this paper shows that the current routers employed in the Internet are clue-less; Namely, it is possible to speedup the IP-lookup by an order of magnitude without any major changes to the existing protocols.