F. Hao and E. W. Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation," in IEEE INFOCOM, 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....of MetaNet channels, the bandwidth tracker has to perform a continuous aggregation of the network topologies that make up the MetaNet. In the literature, there are three typical approaches to topology aggregation: the Symmetric Node method, the Full Mesh method, and the Star method (e.g. [AwD98, HaZ00, IwS98, Lee95]) The symmetric node approach collapses the entire topology into one virtual node and advertises one parameter that reflects the properties of the original topology. In the Full Mesh approach, a logical link connects each of the nodes in the network to every other node. Each virtual link has its ....

F. Hao and E. Zequra, "On scalable QoS routing: Performance evaluation of topology aggregation," IEEE INFOCOM '00, Vol. 1, March 2000, pp. 147-156 .

.... routing [22] dynamic routing [45] premium class routing [46] ticketbased probing [7] Hierarchical Most Reliable Path (MRP) 21] Multiple traffic classes Hop by hop QoS routing for best effort flows [36] optimal premium class routing [46] Scalability General Topology aggregation [47] Hop by hop Selective Ticket based probing [7] Hierarchical Partitioning QoS requirements [48] PNNI [9] rithms are designed to satisfy bandwidth, delay, jitter, and cost constraints. The time complexity of current proposals is generally polynomial. Table VII classifies current proposals, which ....

.... small (less than 1 ) In order to scale better, only aggregate information is advertised outside a domain (recall Figure 2) Topology aggregation mechanisms may not always negatively impact routing performance, depending on the QoS routing algorithm, network topology, and state update frequency [47]. An example of aggregation with widest shortest path routing is provided in [47] In this work, hop count information is advertised infrequently but in detail, while the available bandwidth is advertised more frequently but in less detail. An alternative approach limits QoS routing decisions to ....

[Article contains additional citation context not shown here]

F. Hao and E. Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation," in Proceedings of the Conference on Computer Communications (IEEE INFOCOM '00), March 2000.

....Hierarchical routing and topology aggregation can reduce the protocol overhead. The neighboring nodes that do not receive detailed information receive aggregated information. The two aggregation methods that can be considered suitable for QoS routing are hybrid aggregation and weighted aggregation [40]. Hybrid aggregation can be used for widest shortest path routing 6 that advertises hop count information less frequently and in more detail while advertising available bandwidth more frequently and in more detail. This is based on the fact that hopcount changes much less frequently than available ....

Fang Hao, Ellen W. Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation", in IEEE INFOCOM 2000, pp. 147-156.

.... communication networks may no longer permit flows to be routed based on exact network state information, either due to scalability or routing overhead limits [6] Researchers have proposed various ways to aggregate network topology accompanied by hierarchical routing to resolve this problem, see [3, 8] for examples. With the need to support a mixture of both local and transit flows which would presumably see the exact and abstracted network states respectively, we propose a novel hierarchical shortest path routing scheme based on a consistent routing metric over aggregated abstracted resources. ....

....is to find a representation of capacity availability (shortness) between border nodes. The key abstraction of capacity availability between a border node pair is to find a balance among the number of possible routes connecting the pair, the number of links each route traverses, and 1 See [3, 8, 6] for a detailed description of hierarchical network models. 4 1 1.5 2 2.5 3 3.5 4 4.5 5 0.5 0.6 0.7 0.8 0.9 1 Traffic Intensity Routed Volume The Performance of High Level CAC CAC exp no cac exp CAC bimodal no cac bimodal CAC pareto no cac pareto Figure 2: Performance ....

F. Hao and E. W. Zegura. On scalable qos routing: Performance evaluation of topology aggregation. IEEE INFOCOM

....Pseudo code for processing the Flooding message. 5. Simulation Results This section evaluates the performance of QHMRP through simulation under different conditions. The studies are performed in two stages. In the first stage, a network with 56 nodes and 77 links is used in the simulation study [8]. The network is organized hierarchically into domains. A controller is specified for each domain. The bandwidth capabilities of all links in 1 domain are 155 Mbps and all other links are 622 Mbps. The background traffic load on each link is randomly generated from the range [0, 155] Mbps for ....

F. Hao and E. W. Zegura, On Scalable QoS Routing: Performance Evaluation of Topology Aggregation, Technical Report GIT-CC-99-04, College of Computing, Georgia Tech, 1999.

....blocking probability, routing stability and overhead. 4.1 Simulation Environment Figure 9 shows the two topologies, isp and rand, used in our study. The isp topology of an ISP backbone network is also used in [1, 20] The rand topology is a random graph generated by GT ITM [36] and used in [37]. For simplicity, all the links are assumed to be bidirectional and of equal capacity in each direction. The rand topology has three types of links: thin, thick and dotted while isp topology has only thin links. All thin links have same capacity with C 1 units of bandwidth and similarly all the ....

Fang Hao and Ellen Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation", IEEE INFOCOM 2000.

....As the network becomes larger and larger, it is impossible to broadcast the whole topology to every node in the network to do routing, because it takes enormous amount of bandwidth, time and space. One approach to reduce the bandwidth, time and space complexity is topology aggregation [5]. In the ATM Forum, PNNI [2] standard proposed a possible topology aggregation protocol. In this protocol, nodes are grouped hierarchically into several clusters called peer groups. Although PNNI defines how the aggregated peer group looks like, it does not specify how to do the aggregation. Some ....

F. Hao, E. Zegura. On Scalable QoS Routing: Performance Evaluation of Topology Aggregation. In IEEE Proceedings of the INFOCOM'00, Mar. 2000.

....total connection time of all successful join requests total number of join requests 44 calculating the message overhead, sending a Flooding message over a link is counted as one message. The network used in the simulation consists of 56 nodes and 77 links, and is presented in Figure 9 [HaZ99]. The figure also shows the domains and sub domains used to organize the network into the hierarchical structure. The controller for each domain is specified in the simulation. All the routers are given unique IDs according to the naming scheme described in Section 3.2. The bandwidth capabilities ....

F. Hao and E. W. Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation," Technical Report GIT-CC-99-04, College of Computing, Georgia Tech, 1999.

....topology of a domain is then aggregated to show only the cost of routing through the domain. Each node has a detailed topology of the domain it belongs to, but only an aggregated topology information about other domains. The process of aggregating a network topology is called topology aggregation. [1] shows that topology aggregation reduces This work was supported by the Airforce grant under contract number F30602 97 2 0121 and the National Science Foundation Career grant under contract number NSF CCR 96 23867. the routing overhead by orders of magnitude and does not always have a negative ....

F. Hao, E. Zegura. On Scalable QoS Routing: Performance Evaluation of Topology Aggregation. In IEEE Proceedings of the INFOCOM'00, p. 147-156, Mar. 2000.

....level, topology aggregation is introduced to reduce the amount of information a router needs to maintain so that the scalability can be achieved. The idea is to aggregate each domain into a simple topology, e.g. a star, which captures the cost of going from one end of the domain to another [12] [8]. This simplified topology is broadcasted and is used in finding an inter domain path. At the intra domain level, many QoS routing protocols have been proposed recently, and they can be grouped into two major classes: source routing and distributed routing [3] In source routing, each node ....

Fang Hao, Ellen Zegura. On Scalable QoS Routing: Performance Evaluation of Topology Aggregation. In IEEE Proceedings of the INFOCOM'00, Mar. 2000.

....As the network becomes larger and larger, it is impossible to broadcast the whole topology to every node in the network to do routing, because it takes enormous amount of bandwidth, time and space. One approach to reduce the bandwidth, time and space complexity is to do topology aggregation [6]. In the ATM Forum, PNNI [2] standard proposed a possible topology aggregation protocol. In this protocol, nodes are grouped hierarchically into several clusters, called domains or peer groups. Although PNNI de nes how the aggregated peer group looks like, it does not specify how to do the ....

F. Hao, E. Zegura. On Scalable QoS Routing: Performance Evaluation of Topology Aggregation. In IEEE Proceedings of the INFOCOM '00, Mar. 2000.

....port to port distances. Hybrid aggregation. Note that we have more choices for aggregation when the path cost metric consists of more than one value. Specifically, we have proposed a hybrid aggregation approach for widest shortest routing, which is a combination of full mesh and star aggregation [11] The idea comes from the following observation: hop count information changes much less frequently than available bandwidth information. Only a significant bandwidth shift that changes the path feasibility causes hop count to change. Thus we can afford to update hop count information less ....

....remains better than static, non QoS based routing when appropriate parameters for change based triggers are selected [2] 4. 2 Routing algorithms We have shown that different routing algorithms have different sensitivity to change in routing update interval, or value of clamp down timers [11] Different routing algorithms give different priority to minimizing hop count and balancing traffic load. Widest shortest routing, which gives minimizing hop count the highest priority, is the most insensitive to change in routing update intervals, and has best performance when the update ....

[Article contains additional citation context not shown here]

F. Hao and E. W. Zegura. On scalable QoS routing: performance evaluation of topology aggregation. Technical Report GIT-CC-99-04, College of Computing, Georgia Tech, 1999. Submitted to IWQoS'99.

No context found.

F. Hao and E. W. Zegura, "On Scalable QoS Routing: Performance Evaluation of Topology Aggregation," in IEEE INFOCOM, 2000.

No context found.

F. Hao and E. W. Zegura. On Scalable QoS Routing: Performance Evaluation of Topology Aggregation. In Proc. of IEEE INFOCOM, Tel Aviv, Israel, Mar 2000.

No context found.

F. Hao, and E. W. Zegura. On scalable QoS routing: performance evaluation of topology aggregation. In Proceedings of INFOCOMM'2000.

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