Fig 2 - uploaded by Mehrdad Nourani
Content may be subject to copyright.
Source publication
We present a search algorithm implementable by a parallel architecture based on partitioned forwarding table. This scheme effectively reduces the complexity of finding “the longest prefix match” problem to “a prefix match” problem. The main feature of this algorithm is twofold. First, it advocates intelligent partitioning of the forwarding table to...
Context in source publication
Similar publications
This memorandum outlines the goals for and constraints on an IP address and prefix autoconfiguration mechanism for mobile ad hoc networks.
We propose a charging scheme for cost distribution along a multicast tree when cost is the responsibility of the receivers. This scheme focuses on QoS considerations and it does not depend on any specific type of service. The scheme has been designed to be used as a bridge between unicast and multicast services, solving the problem of charging mult...
This paper describes simulations of speech se- quences transmission for intrusive measurement of voice transmission quality of service (VTQoS) in the environment of IP networks. The aim of the simulations was to investigate the impact of the different durations of speech sequences on speech quality from the jitter rate and packet loss point of view...
Multimedia service provisioning in Packet-Switched networks is one of the most active research and standardization efforts nowadays, and constitutes the most likely evolution of current telecommunication networks. In this environment, the 3GPP IP Multimedia Subsystem (IMS) represents the service provisioning platform of choice for SIP-based content...
There is a common misconception that internal IP networks (intranets) are secure; and that external networks such as the Internet are vulnerable and unsecured. The truth of the matter is that there is little on a network that is truly secure. The very nature of devices being networked makes them vulnerable on an IP network to various attacks and sp...
Citations
... Several algorithms have been presented in the sense that a binary search tree composed of completely disjointed entries is perfectly balanced. In IP packet forwarding based on partitioned lookup table (IFPLUT) [16], a partitioning technique of multiple tables has been suggested for creating a set of disjointed prefixes. Using the fact that prefixes with the same output port are mutually disjointed, all entries are partitioned according to their output port. ...
Due to a tremendous increase in internet traffic, backbone routers must have the capability to forward massive incoming packets at several gigabits per second. IP address lookup is one of the most challenging tasks for high-speed packet forwarding. Some high-end routers have been implemented with hardware parallelism using ternary content addressable memory (TCAM). However, TCAM is much more expensive in terms of circuit complexity as well as power consumption. Therefore, efficient algorithmic solutions are essentially required to be implemented using network processors as low cost solutions.
... [12] ...
The major challenge to design and deployment of mobile ad hoc networks (MANETs) is its dynamic nature, which carries with itself a set of security measures to be resolved. In this paper, we compare the behavior of three routing protocols DSDV, DSR and AODV under security attack, where the investigation is carried out with respect to two types of node misbehavior. The parameters taken into consideration for evaluation of network performance are normalized throughput, routing overhead, normalized routing load and average packet delay, when a certain percentage of nodes misbehave. It could be established through simulation results that DSDV is the most robust routing protocol under security attacks as compared to the other two. In addition, it reveals that a proactive routing protocol reduces the impact of security attack by excluding the misbehaving nodes in advance.
... [12] ...
Web server logs stores click stream data which can be useful for mining purposes. The data is stored as a result of user's access to a website. Web usage mining an application of data mining can be used to discover user access patterns from weblog data. The obtained results are used in different applications like, site modifications, business intelligence, system improvement and personalization. In this study, we have analyzed the log files of smart sync software web server to get information about visitors; top errors which can be utilized by system administrator and web designer to increase the effectiveness of the web site.
... [12] ...
The growth of the World Wide Web (WWW) has seen it evolve into a rich information resource. It is constantly traversed with the aid of crawlers so as to harvest web content. When collecting data, crawlers have the potential of causing service denial to web servers. This paper proposes the application of sampling as a selection strategy in the design of structural analysis web crawlers. This has the benefit of alleviating the problems of bandwidth costs to web servers whilst retaining the quality of the data that is mined by crawlers. The initial results of this study are promising and are presented in this paper.
... [12] ...
So that the routers forward an IP packet with his destination, they are running a forwarding decision on an incoming packet to determine the packet's next-hop router. This is achieved by looking up the longest matching prefix with a packet destination address in the routing tables. Therefore, one major factor in the overall performance of a router is the speed of the IP address lookup operation due to the increase in routing table size, in this paper, a new IP address lookup algorithm based on cache routing-table is proposed, that contains recently used IP addresses and their forwarding information to speed up the IP address lookups operation in the routers. We evaluated the performance of our proposed algorithm in terms of consultation time for several sets of IP addresses, the results of performance evaluation show that our algorithm is efficient in terms of the lookup speed since search can be immediately finished when the input IP address is found in the cache routing table.
... Many approaches have been used to improve the scheduler performance. They include efficient lookup algorithms [18], better routing table designs [19], hardware-based lookup methods [20], and elimination of the lookup process by using self-routing schemes [21]. ...
For efficient utilization of bandwidth in optical packet switching,
the guard time ${T}_{\rm
g}$ between packets should only be a small fraction of the
packet transmission time ${
T}_{\rm d}$. Since the guard time ${ T}_{\rm g}$ of existing packet switching
approaches must be larger than the reconfiguration time ${T}_{\rm sw}$ of optical switches, this
imposes a stringent demand on the switch reconfiguration time ${T}_{\rm sw}$ as the transmission rate
of optical fibers increases. By using batch transfer of packets or multiple
switching fabrics in parallel, the requirement on the switch reconfiguration
time can be significantly relaxed. The utilization of the transmission links
can be greatly improved because the guard time between packets is no longer
constrained by the switch reconfiguration time.
... Several software and hardware approaches for IP address lookup problem have recently been proposed [4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22], and [28][29][30][31][32]. Binary trie which is the base of many existing schemes [4] is a binary tree which is searched starting from the root and recursively going down to the left or right child depending on the current bit of the searched address. Any time a prefix is seen at a node, it is returned as the longest matched prefix up to now. ...
... However, this method suffers from unbalanced distribution of prefixes over partitions. This problem was addressed in [15] by dynamically allocating memory pools to partitions. ...
... Our architecture 226,094 193 Reconfigurable BDD [21] 33,796 157.7 Partitioned table [15] 18,684 68 Reconfigurable FSM [20] 38,367 22.22 LogW-Elevators (software) [31] 33,796 20.5 FIPL (parallel trie bitmap) [30] 16,564 9.09 1 A frame is a column of logic resources inside the FPGA. Table 6 The results of the amount of memory usage and the number of memory accesses per lookup in our proposed architecture and other approaches [11] 2000 29,584 67.6 1 DIR-24-8 [22] 264,000 --1-2 Trie bitmap [28] 20,000 41,811 478.34 1-5 Indirect lookup [29] 3600 40,000 90 1-3 Parallel hashing [17] 1512 37,000 40.86 1-5 Multi-way search [7] 5600 30,000 186.67 1-9 DTBM [32] 36,000 85,987 418.67 1-12 EnBiT (software) [12] 2413 41,000 58.86 parameters. ...
With today’s networks complexity, routers in backbone links must be able to handle millions of packets per second on each of their ports. Determining the corresponding output interface for each incoming packet based on its destination address requires a longest matching prefix search on the IP address. Therefore, IP address lookup is one of the most challenging problems for backbone routers. In this paper, an IP routing lookup architecture is proposed which is based on a reconfigurable hardware platform. Experimental results show that the rate of 193 million lookups per second is achieved using our architecture while prefixes can be updated with a rate of 3 million updates per second. Furthermore, it was shown that using our reconfigurable architecture results in rare update failure rate due to resource limitations.
... IP packet Forwarding based on Partitioned LookUp Table (IFPLUT) is work by partitioning the Forwarding Table (FT) based on the egress port [13][14] [15]. IFPLUT focus only on a unicast (single-source, single-destination) routing with the following assumptions -Suppose P = {p1 , p2 , … , p G } is the set of G prefixes collected by a router. ...
... So the algorithm break the one-to-one correspondence between PFTs and partitions and have a big pool of search units and let a management unit dynamically assign them to partitions whenever the need exists. The IFPLUT algorithm was enhanced in [13] by using a pool of distinct memory units. Each partition (corresponding to an egress port) will be assigned zero or more memory units. ...
... So, the minimum memory lost will happen when K=G+1. (11) By simplifying (11) we get ( ) 1 , max min N W = (12) And of course every router has more than one interface so, the minimum memory lost will be N W = min (13) And now the maximum memory loss can be happen at K=N+1, so (14) Simplifying (14) yield to ...
In contrast to the one-to-one model of IP unicast, in which data packets are sent from a single source to a single recipient, IP multicast provides a method of efficient many-to-many communication. This concept is becoming increasingly important, both in the Internet and in private networks, for providing services such as multimedia content delivery. IP packet Forwarding based on Partitioned Look Up Table (IFPLUT) is an algorithm used to partition the Forwarding Table (FT) based on the egress port. IFPLUT focus only on a unicast routing (single-source, single-destination). We present in this works a modified IFPLUT hardware-based algorithm to support multicast routing. Moreover a hardware architecture design is proposed for the modified algorithm with a study to optimize the routing memory loss.
... A straightforward VLSI implementation of these components indicate that [17]: t rcam block t tcam block t priority encoder >> t encoder (7) Therefore, by combining Equations 5, 6 and 7, we can simplify the difference in search speed between the RCAM and TCAM architectures as follows: ...
A novel reconfigurable content addressable memory, called RCAM, is proposed that supports on-the-fly reconfiguration between CAM and TCAM. The area overhead of the proposed RCAM cell is only 5.6% when compared to conventional TCAM. This overhead is compensated by area saving due to removal of the priority encoder. Other features of our architecture include reconfigurability, and better overall performance and power. To achieve these we incorporated two novel techniques: (i) a hybrid CAM/TCAM architecture that allows user to pre-define CAM/TCAM cell behavior in each bit or word position and ultimately curtail the overall power consumptions of memory unit: and (ii) a wired-AND technique by which we can completely eliminate the sorting requirement and thus significantly reduce the update time. A 4 Kb RCAM architecture was implemented using 0.18 mum CMOS technology. The simulations indicate a search time of 6.15 ns, i.e. capability of handling about five OC-192 at wire speed.
... Moreover, prefix F did not get any chance to be examined in the process of the binary search. By carefully analyzing the binary search on the sorted prefixes, we found that the main cause of the failure comes from the enclosure property between prefixes of different lengths [23,9]. In the above example, prefix O is enclosed by F. The search failure for address 01011000 is caused by the absence of a prefix that covers 01011000 and is also smaller than prefix O. ...
Backbone routers with tens-of-gigabits-per-second links are indispensable communication devices to deploy on the Internet. The IP lookup operation is the most critical task that must be improved in routers. In this paper, we first present a systematic method to compare prefixes of different lengths. The list of prefixes can then be sorted and stored in a sequential array, which is contrary to the linked lists used in most of trie-based structures. Next, fast binary and multiway prefix searches assisted by auxiliary prefixes are proposed. We also developed a 32-bit representation to encode the prefixes of different lengths. For the large routing tables currently available on the Internet, the proposed multiway prefix search can achieve the worst-case number of memory accesses of three and four if the sizes of the CPU cache lines are 64 bytes and 32 bytes, respectively. The IPv4 simulation results show that the proposed prefix searches outperform the existing IP lookup schemes in terms of lookup times and memory consumption. The simulations using IPv6 routing tables also show the performance advantages of the proposed binary prefix searches. We also analyze the performance of the existing lookup schemes by concurrently considering the lookup speed, the update speed, and the memory consumption. Although the update speed of the proposed prefix search is worse than the dynamic routing table schemes with log(N) complexity for a table of N prefixes, our analysis shows that the overall performance of the proposed binary prefix search outperforms all the existing schemes.
