L. Kleinrock and F. Kamoun. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1:155-174, 1977.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....has been developed, including prefix routing schemes [13] boolean routing schemes [40] multidimensional interval routing schemes [41,109] and more. For surveys of the many developments in this area see [125,55] The problem of efficiency memory tradeoffs for routing schemes was first raised in [76], which proposed the general approach of hierarchically clustering a network into levels and using the resulting structure for routing. The total memory used by the scheme is O(n Delta log n) However, the method of [76] is based on making some fairly strong assumptions regarding the ....

....of efficiency memory tradeoffs for routing schemes was first raised in [76] which proposed the general approach of hierarchically clustering a network into levels and using the resulting structure for routing. The total memory used by the scheme is O(n Delta log n) However, the method of [76] is based on making some fairly strong assumptions regarding the existence of a certain partition of the network, and moreover, the method does not provide an algorithm for computing such a partition if it exists. Several variations and or improvements were studied later, cf. 77,104,114] Most ....

[Article contains additional citation context not shown here]

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks; performance evaluation and optimization. Computer Networks, 1:155--174, 1977.

....in part by the National Science Foundation under NSF CAREER grant ANI 0132802. The autonomous set up of addressing and routing in a large network requires the decomposition of the network into connected clusters. Bounding the cluster size is a requirement imposed by routing protocol complexity [5], 6] 8] 10] Furthermore, resource constraints or specific network architectures can impose limits on the cluster size. Thus, it is desirable to develop distributed algorithms that can form bounded size clusters. Moreover, the cluster sizes produced should be as close as possible to the ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, Volume 1, Number 1, pp. 155--174, 1977.

....complete, it provides the backbone and several non trivial algorithmic solutions. In our upcoming implementation, we expect to complete and finetune our design. Furthermore, we expect this work to provoke a constructive reevaluation of current networking architectures. Related work. Area Routing [11] and Landmark routing [16] are classical papers on hierarchical routing. LANMAR [13] and L [9] are modern extensions of Landmark routing, whereas PeerNet to our knowledge is the first protocol for dynamic networks with similarity to Area Routing. For a survey of ad hoc routing, see [10] and for ....

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization, . Computer Networks, 1, 1977.

....use of (a, t) Clusters in ZRP could achieve more efficient and adaptive hybrid routing without significantly increasing its complexity. 3 (a,t) Cluster Framework Hierarchical routing has been shown to be essential in order to achieve at least adequate levels of performance in very large networks [17, 18]. In fixed infrastructure networks, hi erarchical aggregation achieves the effect of making a large network appear much smaller from the perspective of the routing algorithm. Cluster based routing in ad hoc networks can also make a large network appear smaller, but, more importantly, it can make ....

F. Kamoun and L. Kleinrock. Hierarchical routing for large networks: performance evaluation and optimization. Computer Networks, (1977.

....to make dynamic routing more scalable, one potential solution is hierarchical clustering and routing. The first hierarchical routing proposal was by McQuillian [2] The characteristics and performance of the hierarchical approach were later evaluated and optimized by F.Kamoun and L. Kleinrock in [3]. Their scheme was designed to minimize storage of routing tables, and to propagate and update them very effectively without introducing significant performance degradation. Drawing from the work in [2] 3] C.Alaettinoglu proposed a routing scheme based on the distributed Bellman Ford algorithm ....

....the hierarchical approach were later evaluated and optimized by F.Kamoun and L.Kleinrock in [3] Their scheme was designed to minimize storage of routing tables, and to propagate and update them very effectively without introducing significant performance degradation. Drawing from the work in [2] [3], C.Alaettinoglu proposed a routing scheme based on the distributed Bellman Ford algorithm [6] and Ramamoorthy et al. 4] 5] proposed a scheme based on link state algorithm. The most recent effort in hierarchical clustering and routing is Private Network Network Interface (PNNI) routing [7] a ....

[Article contains additional citation context not shown here]

L. Kleinrock and F.Kamoun. "Hierarchical routing for large networks - Performance evaluation and optimization", Computer Networks, Vol.1, No. 3, pp.155-174, Jan. 1977

....algorithm etc. These flat routing scheme can easily overload the channel capacity by sending nothing but large size periodical routing table updating messages among the nodes when the size of network becomes large. Hierarchical techniques are commonly used in wired network for scalability [11, 13, 21, 24, 25]. For wireless networks, a hierarchical clustering and routing scheme based upon physical location management was recently proposed in [6, 12] This scheme, however, creates implementation problems which are potentially complex to resolve. First, it does allocate Cluster IDs dynamically. This ....

L. Kleinrock and F.Kamoun. Hierarchical routing for large networks - performance evaluation and optimization. Computer Networks, 1(3):155--174, January 1977.

....LVAs report links needed to reach destinations, not complete topology maps, and a destination can be a single entity or an aggregate of entities. Therefore, aggregation of information can take place in an LVA by adapting any of the area based routing techniques proposed for DVAs in the past (e.g. [Gar88, KK77, Tsu88]) The following sections introduce the network model used in the remainder of this thesis, describe LVA, show that LVA converges to correct paths a finite time after the occurrence of an arbitrary sequence of link cost or topological changes under the assumption that all routers run the same ....

....must be routed via the backbone. 80 There have been many hierarchical routing proposals described in the past based on the notion of areas, which are also called clusters [Ste95] The first such proposal was McQuillian s [McQ74] this proposal was analyzed in detail by Kamoun and Kleinrock [KK77]. Most prior proposals on hierarchical routing have routing algorithms based on topology broadcast or variations of the distributed Bellman Ford algorithm. Ramamoorthy et al. RT83, TRTN89] proposed an algorithm based on link state information for hierarchical routing. According to this algorithm, ....

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1(3):155--174, 1977.

....backbones, the dissemination of complete cluster topology within a cluster, or the dissemination of the complete inter cluster connectivity among clusters. Furthermore, ALP can be used with distributed hierarchical routing schemes proposed in the past for both distance vector or link state routing [32, 10, 38, 2]. Because routers in ALP propagate link state information selectively, it incurs less communication overhead than algorithms based on topology broadcast. The following sections introduce the network model assumed throughout the rest of the chapter, describe ALP, show that ALP converges to correct ....

....to all its neighbors incrementally. The aggregation of adjacent links and source trees reported 40 by neighbors constitutes the partial topology known by a router. STAR can be used with distributed hierarchical routing schemes proposed in the past for both distance vector or linkstate routing [32, 10, 37, 2]. Prior proposals for link state routing using partial link state data without clusters [19, 20] require routers to explicitly inform their neighbors which links they use and which links they stop using. In contrast, because STAR sends only changes to the structure of source trees, and because ....

L. Kleinrock and F. Kamoun. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1:155-174, 1977.

....AS overlay may be able to explain some of the observed macroscopic properties of the inter AS topology in terms of the underlying physical structure. To place our work in context, we point out two important caveats. First, it is well known that hierarchical routing can result in non optimal paths [12]. Our paper quantifies the extent to which hierarchical routing in the Internet, together with routing policy, affects paths. Second, the correlation, if any, between path length and end to end delay is poorly understood. As such, then, our results cannot be directly extrapolated to observed ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, vol. 1, pp. 155--174, 1977.

....backbones, the dissemination of complete cluster topology within a cluster, or the dissemination of the complete inter cluster connectivity among clusters. Furthermore, STR can be used with distributed hierarchical routing schemes proposed in the past for both distance vector or link state routing [25, 24, 6, 26]. The only prior proposals for link state routing using partial topology information without clusters are the adaptive link state protocol (ALP) 27] and the link vector algorithm (LVA) 28] These protocols differ primarily on the way in which routers tell one another which links they stop using ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, Vol. 1, pp. 155-174, 1977.

....Hierarchy, however, by itself does not present a satisfactory solution. Therefore, we discuss our assumptions about locality in traffic patterns, and how we exploit this locality to control the overhead of routing information distribution. 3. 1 Hierarchical Routing In hierarchical routing [17], network nodes are grouped into larger aggregates, or clusters. Detailed information about individual nodes in a cluster is distributed only within the cluster. External advertisements hide details about the individual nodes, and nodes in one cluster cannot distinguish among the elements in other ....

Leonard Kleinrock and Farouk Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1:155--174, 1977.

....links needed to reach destinations, not complete topology maps, and a destination can be a single entity or an aggregate of entities. Therefore, aggregation of information can take place in an LVA by adapting any of the area based routing techniques proposed for DVA s in the past (e.g. 8] [19], 29] The following sections introduce the network model assumed throughout the rest of the paper, describe LVA, show that LVA converges to correct paths a finite time after the occurrence of an arbitrary sequence of link cost or topological changes under the assumption that all routers run ....

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization," Computer Networks, vol. 1, pp. 155--174, 1977.

....Efficient, Algorithms for Self structuring Large Networks, project supported by DARPA under contract DABT63 96 C 0100. Hierarchical structuring as a means to improve the performance of routing in large networks was first introduced by McQuillan [27] and first analyzed by Kamoun and Kleinrock [21], 16] Many others have since proposed hierarchical routing procedures [9] 4] 29] 34] 6] two of which Internet OSPF [28] and ATM PNNI [2] have been implemented and widely deployed. These approaches have been designed primarily for fixed networks in which organization of the ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, Vol. 1, No. 1, pp. 155-174, 1977.

....of distributed agents, since it requires configuration of nodes which are not part of the hierarchy. It is, however, applicable when the same policy constraints apply to multiple protocols, so that similar hierarchies are desirable. 2. 3 Area hierarchies A node s address in an area hierarchy [11 13] also corresponds to its position in the hierarchy. In an area hierarchy, however, a node is only aware of its children, its siblings, and the siblings of its direct ancestors; it is not necessarily aware of other nodes which are topologically close. Hagouel [13] and Shacham [12] both discuss ....

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1:155--174, 1977.

....given the lack of previous work on the specific topic of this paper. The most relevant body of work to our problem of QoS routing in the presence of inaccurate information, is a set of papers aimed at exploring state aggregation issues and their impact on routing performance in large networks. [16] provided an initial investigation of the problem of hierarchical routing in large networks, and focused on the problem of cluster design to ensure maximum reduction of routing table sizes without substantial increases in path length (hop count) 17] focused on the problem of state aggregation in ....

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks - performance evaluation and optimization," Computer Networks, vol. 1, pp. 82--92, 1977.

....in ZRP could achieve more efficient and adaptive hybrid routing without significantly increasing its complexity. 3 (ff; t) GammaCluster Framework Hierarchical routing has been shown to be essential in order to achieve at least adequate levels of performance in very large networks [17, 18]. In fixed infrastructure networks, hierarchical aggregation achieves the effect of making a large network appear much smaller from the perspective of the routing algorithm. Cluster based routing in ad hoc networks can also make a large network appear smaller, but, more importantly, it can make a ....

F. Kamoun and L. Kleinrock. Hierarchical routing for large networks: performance evaluation and optimization. Computer Networks, (1977)(1):155--174, 1977.

....the dissemina tion of complete cluster topology within a cluster, or the dissemination of the complete inter cluster connectivity among clusters. Furthermore, STAR can be used with distributed hierarchical routing schemes proposed in the past for both distance vector or linkstate routing [16, 18, 22, 9]. Prior proposals for link state routing using partial link state data without clusters [13, 14] require routers to explicitly inform their neighbors which links they use and which links they stop using. In contrast, because STAR sends only changes to the structure of source trees, and because ....

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1:155-- 174, 1977.

....traffic must be routed via the backbone. There have been many hierarchical routing proposals described in the past based on the notion of areas, which are also called clusters [7] The first such proposal was McQuillian s [8] this proposal was analyzed in detail by Kamoun and Kleinrock [9]. Most prior proposals on hierarchical routing have routing algorithms based on topology broadcast or variations of the distributed Bellman Ford algorithm (e.g. 10] 11] More recently, Murthy and Garcia Luna Aceves [12] proposed an area based hierarchical routing algorithm called HIPR that is ....

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization," Computer Networks, vol. 1, no. 3, pp. 155--174, 1977.

....identifies the cluster of subnets that belong to a given network. In a hierarchical locator, the lowest level (0th level) component is the host (or router) The next higher level (1st level) component is a cluster of hosts. The 2nd level component is a cluster of 1st level clusters, and so on [62]. This thesis refers to both hosts and clusters as hierarchy elements. The topological constraint placed on clusters is that there must be a path between any two nodes in a cluster that only traverses nodes that belong to the cluster. In other words, it must be possible for nodes outside a ....

....more complex, because the number of up choices increases. 3.1.2 Mesh Topology It is possible to form a cluster hierarchy even when the topology displays no hierarchical characteristics. All that is necessary is to logically group hierarchical elements such that each group is internally connected [62]. Figure 3.4A shows a mesh topology. Figure 3.4B shows two levels of clusters superimposed on the mesh topology. It also shows the node addresses that result from this clustering. As stated before, the only hard requirement for this clustering is that each cluster be internally connected. There ....

[Article contains additional citation context not shown here]

F. Kamoun and L. Kleinrock. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1(3):155--174, January 1977.

....stability, routing in large networks, and loop free routing. Specific issues regarding routing in mobile wireless environments are introduced in section 4. The content of these sections reflect the substantial contribution to the literature in the field, of a select group of papers: 2] 21] [14] [15] 7] 4] 9] 8] 10] 16] 3.2 The Stability Problem Routing stability refers to the property that a routing algorithm adapts smoothly to network dynamics without suffering wild oscillations. Within a finite time after any resource failure, change in traffic conditions or arbitrary change ....

....from local dynamics, limit the possibility of widespread instability, and speed convergence after significant changes. While McQuillin [17] was among the first to suggest the importance of hierarchical clustering and routing information aggregation, it was Kleinrock and Kamoun s classic work [14] [15] that established the basis for systematically analyzing and quantifying the structure and performance of hierarchical routing. Substantially all subsequent research on the topic draws upon their results. 3.3.1 Analysis of an m level Hierarchical Clustering Scheme There are three basic ....

[Article contains additional citation context not shown here]

Leonard Kleinrock and Farouk Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, (1977)(1):155--174, 1977.

....or hybrid routing techniques. The idea of dynamic cluster based routing, and its variations, was originally conceived as a means of achieving network scalability [63, 1] in PR networks. Early adaptations were intended to achieve to same objectives as hierarchical routing in fixed networks [36]. However, unlike the cluster organization of a fixed network, the assignment of nodes to clusters in PRNETs and ad hoc networks must be a dynamic process. In a clustered ad hoc network, the network is dynamically organized into partitions called clusters with the objective of maintaining a ....

....strategy must be implemented to setup inter cluster routes. 3. 6 (ff; t) GammaCluster Algorithm Previous research has shown that hierarchical routing, which is a special case of cluster based routing, is essential in order to achieve at least adequate levels of performance in very large networks [36, 37]. In fixed infrastructure networks, hierarchical aggregation achieves the effect of making a large network appear much smaller from the perspective of the routing algorithm. Cluster based routing in ad hoc networks can also make a large network appear smaller, but, more importantly, it can make a ....

F. Kamoun and L. Kleinrock. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, (1977)(1):155--174, 1977.

....been used for information hiding in computer networks. Researchers in the area of routing have used hierarchical clustering to reduce the information stored in the network nodes. The reduction of routing information in large networks is realized through a hierarchical clustering of network nodes [21]. In the following sections we describe two clustering techniques that have been used in routing. 4.3.1 Area Hierarchy Area hierarchy is being widely used for routing in large networks. The entire network is organized as a tree of areas, in which each area at a higher level is made up of one or ....

....traffic in networks is generated based on the state stored in the nodes. The intent of this scaling mechanism is to limit the growth of the volume of the control traffic by reducing the information stored in the network. Unicast routing protocols use aggregation to reduce the routing table sizes [21]. The main idea is for a particular router to maintain complete routing information about nodes that are close to it, and less information about nodes located further away from it. Reduction in the size of the routing tables also reduces the volume of the control traffic generated by routing ....

[Article contains additional citation context not shown here]

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1:155--174, 1977.

....of the five evaluation criteria. Finally, Section X summarizes the characteristics and discusses their pros and cons in the current internet environment. II. Background on Hierarchical Address Assignment Hierarchical routing and addressing is based on the principal of hierarchical clustering [7]. A cluster is a group of network elements connected together such that there is a path from any network element in the cluster to any other network element in the cluster that traverses only network elements in that cluster. A network element can be a single physical component such as a host or ....

F. Kamoun and L. Kleinrock. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1(3):155--174, January 1977.

....report links needed to reach destinations, not complete topology maps, and a destination can be a single entity or an aggregate of entities. Therefore, aggregation of information can take place in an LVA by adapting any of the area based routing techniques proposed for DVAs in the past (e.g. 8] [18], 28] The following sections introduce the network model assumed throughout the rest of the paper, describe LVA, show that LVA converges to correct paths a finite time after the occurrence of an arbitrary sequence of link cost or topological changes under the assumption that all routers run the ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, vol. 1, pp. 155-174, 1977.

.... rather than path characteristics, this approach reduces the combinatorial explosion incurred with any type of DVA for routing under multiple constraints [23] Aggregation of information can take place in an LVA by adapting the areabased routing techniques proposed for DVAs in the past [12] [25], 27] 35] In contrast, aggregation of information in traditional LSAs is difficult, because routers need to define different levels of topologies in order to use topology broadcast methods. The following sections introduce the network model assumed throughout the rest of the paper; describe ....

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, Vol. 1, pp. 155-174.

....each source must send query packets to discover routes. To keep from flooding the network with these queries, we send the queries along paths in C. As a result, the routing becomes similar to hierarchical routing. Hence, the selected routes are not necessarily the shortest paths. As noted in [10], the path lengths increase because routes are constrained to pass through specific nodes. Additional knowledge, gained from local domain topologies, snooping, exchanges of long lived edges, and longcycle completions, can be used to counteract the increase in path length. This extra knowledge ....

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: performance evaluation and optimization, " Computer Networks, 1, (1977) 155--174.

....a similar aggregation process [13] without including Steps 1 and 4. Network partitioning (Step 1) depends on both administrative constraints and performance considerations. Theoretical results have shown that a group size of two or three is optimal in terms of reducing the routing table size [14] But there may be good administrative or routing performance reasons to keep the domains larger. Further research remains to be done to find out the relationship between routing performance and grouping strategy. Exchanging information between domains (Step 4) is similar to exchanging ....

L. Kleinrock and F. Kamoun. Hierarchical routing for large networks: Performance evaluation and optimization. Computer Networks, 1977.

No context found.

L. Kleinrock and F. Kamoun. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1:155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1:155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun. Hierarchical Routing for Large Networks: Performance Evaluation and Optimization. Computer Networks, 1:155--174, 1977.

No context found.

L. Kleinrock and F. Kamoun. "Hierarchical Routing for Large Networks; Performance Evaluation and Optimization ". Computer Networks vol. 1, pp. 155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun. "Hierarchical Routing for Large Networks; Performance Evaluation and Optimization ". Computer Networks vol. 1, pp. 155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical Routing for large networks; performance evaluation and optimization," Computer Networks, Vo 1 , pages 155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization," Computer Networks,vol.1, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization,," Computer Networks,vol.1, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: performance evaluation and optimization," Computer Networks (Elsevier), 1(1):155--174, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization," Computer Networks, vol. 1, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization,," Computer Networks,vol.1, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization," Computer Networks, Vol. 1, pp. 155-174, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: Performance evaluation and optimization," Computer Networks, vol. 1, 1977.

No context found.

L. Kleinrock and F. Kamoun, Hierarchical routing for large networks; performance evaluation and optimization, Computer Networks, 1 (1977), pp. 155-174.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical routing for large networks: performance evaluation and optimization," Computer Networks (Elsevier), 1(1):155--174, 1977.

No context found.

L. Kleinrock and F. Kamoun, "Hierarchical Routing for large networks; performance evaluation and optimization," Computer Networks, Vol. 1, pages 155-174, 1977.

No context found.

Leonard Kleinrock and Farouk Kamoun, Hierarchical routing for large networks Performance evaluation and optimization, Computer networks, pp155-174, 1977

No context found.

L. Kleinrock and F. Kamoun. Hierarchical Routing for Large networks: Performance Evaluation and Optimization. Computer Networks, 1:155--174, 1977.

No context found.

F. Kamoun and L. Kleinrock, "Hierarchical Routing for Large Networks: Performance Evaluation and Optimization, " Computer Networks 1977(1), pp. 155174.

No context found.

L. Kleinrock and F.Kamoun. "Hierarchical routing for large networks - Performance evaluation and optimization", Computer Networks, Vol.1, No. 3, pp.155-174, January 1977

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC