| Y. Breitbart, et al. Topology discovery in heterogeneous IP networks. In IEEE INFOCOM, 2000. |
....cases, each hop can be analyzed independently using the input and output IP addresses for each layer 3 hop. Descriptions of how to perform the layer 2 analysis can be found in [12] using SNMP to obtain information about the 12 switches spanning tree tables) in [13] using active probes) and in [14, 15] (using MIB forwarding tables) 5 Conclusion Finding a layer 3 path between endpoints is an important problem for network analysis. Traditionally, active network probes such as traceroute have been used. These traditional approaches have several short comings. For example, a program must run on ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz, "Topology discovery in heterogeneous IP networks," in Proc. of the
....been proposed (e.g. 5, 6, 7, 8, 9] Despite the importance of the layer 2 topology, little literature is available. One related approach [10] uses pattern matching on interface counters available through SNMP. Another approach to generate the layer2 topology between switches was presented in [11] and improved upon in [12] This approach operates by processing the forwarding tables obtained from each switch via SNMP. Both approaches start by collecting the forwarding tables from each switch in the network. The forwarding table caches entries for each physical address that associate the ....
....entries for each physical address that associate the address with the port toward the host using the physical address. When a packet arrives at the switch, it looks up the destination address in the forwarding table to find the port it should use to forward the packet. The original approach [11] assumes that all physical addresses are cached in every switch (this condition can be forced by sending ping messages between each pair of hosts) They define as the th interface of the th switch and as the set of physical address in the forwarding table for . They prove that two ....
[Article contains additional citation context not shown here]
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz, "Topology discovery in heterogeneous IP networks," in Proc. of the 2000.
....networks since we would ideally like to have better control on retrieved as resolution increases because retrieved overhead directly effects the expended overhead of the process. VI. RELATED WORK Researchers have proposed different mechanisms for topology discovery of data networks [3] 7] [8], 9] primarily using probing techniques. Using routing tables for aggregating topology information is not feasible because traditional routing tables may not be available if data centric model of routing is used [5] Further, in ad hoc deployments, routing tables are often inaccurate or ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri and A. Silberschatz. "Topology Discovery in Heterogeneous IP Networks" In Proceedings of IEEE INFOCOM, 2000.
....network state at multiple resolutions at a proportionate communication cost. The rest of the introduction section provides the motivation, main contributions, overview, and related work. A. Motivation Researchers have proposed different mechanisms for topology discovery of data networks [3] 6] [7], 8] The fundamentally different architecture of sensor networks [1,2] poses new challenges in topology discovery. Using routing tables for aggregating topology information is not feasible because traditional routing tables may not be available if data centric model of routing is used [4] ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri and A. Silberschatz. "Topology Discovery in Heterogeneous IP Networks" In Proceedings ofIEEE INFOCOM, 2000.
....properly upgraded and con gured. The innovations in IP switching equipment in the Nineties require speci c methods for obtaining topology data from them, which are highly nonstandard at this time. Layer 2 discovery is an active area of research, with improvements expected in this dicult area. [4], 38] 2] 5] In summary, topology discovery techniques are improving, but remain a blend of weaker open source contributions and stronger but limited proprietary products. It remains to be seen at what rate the combination of market forces, open source develop 23 ment and standardization ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz. Topology discovery in heterogeneous ip networks. Infocom 2000.
....discovery, and because it is available through a standard SNMP MIB, it is portable enough to enable automatic topology discovery on almost all Ethernet bridges. Recently, Bertsekas et al. described an algorithm for performing topology discovery using information available from Ethernet bridges [3]. Their algorithm obtains good accuracy, but requires that each bridge have a forwarding entry for all other bridges in the network. Because bridges do not normally communicate with each other, and because obtaining complete information from forwarding databases can be a challenging process, this ....
....Assume that F x i and F y j are complete. Two bridges i and j are directly connected via the link connected to port x on i and port y on j if and only if F x i #F y j = 0 and F x i #F y j = N. The proof is omitted here. Interested readers are referred to Breitbart et al. [3] for proof of a similar theorem. 4.1 Shared segments For a simple switched Ethernet, Theorem 4.1 is sufficient, but most networks do not use direct connections to connect all nodes to one another. For example, shared segments are commonly used at the leaves of networks, such as the hub attached ....
[Article contains additional citation context not shown here]
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz. Topology discovery in heterogeneous IP networks. In Proceedings of INFOCOM
....an enterprise network. The work in [4] presents and compares ping , traceroute , and DNS based techniques to obtain the layer 3 topology. Despite the importance of the layer 2 topology, little literature is available. One approach to generate the layer 2 topology between switches was presented in [5] and improved upon in [6] This approach discovers the topology by processing the forwarding tables obtained from each switch via SNMP. Its accuracy depends on how complete forwarding tables are, and it has not been proven with VLANs. Some switch vendors (e.g. 7] have tools for layer 2 topology ....
....layer 2 path can be found between any IP addresses in the same subnet (i.e. between any two devices in the layer 3 path) As with the layer 3 path, the part of the path between network devices is simple the spanning tree defines the path. Finding the first switch in the path is more difficult ([5] explains complexities of this problem) Our approach uses the switch s forwarding tables, the spanning tree, and the router s MAC table, all of which are stored in the database. First, it looks up the endpoint s MAC address from router s MAC table. All switches, except the one closest to the ....
Yuri Breitbart, Minos Garofalakis, Cliff Martin, Rajeev Rastogi, S. Seshadri, and Abraham Silberschatz, "Topology discovery in heterogeneous IP networks," in Proc. of the
....has been surprisingly little research, see e.g. 11] 9] concerning global multicast topology discovery and even less, to our knowledge, concerning local multicast topology discovery. A large amount of work has however been devoted to Internet topology discovery, see e.g. 12] 13] 14] [15]. By contrast with multicast topology discovery, Internet topology information can be collected during long time scales (e.g. several days or even several weeks) 12] or by passive probing [16] since the physical topology remains stable over reasonably long time periods. In the case of ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz, "Topology discovery in heterogeneous IP networks," in Proceedings IEEE Infocom 2000, 2000.
....re organize continuously their structure and therefore their communication network, a particular challenge in con guration management is keeping up with frequent element, topology and policy changes. Reaching this aim is partly facilitated using sophisticated network discovery mechanisms, cf. [8]. However, the discovery mechanisms still require some minimal con guration on the systems, e.g SNMP access should be feasible. Often, this is not possible due to incomplete element installation. The aim of having complete and up to date con guration information can be hardly achieved. Network ....
Y. Breitbart, M. Garofalakis, C. Cli Martin, R. Rastogi, S. Seshadri, and A. Silberschatz, \Topology discovery in heterogeneous ip networks," in Proceedings of the IEEE Infocom 2000, Tel-Aviv. Israel, 2000, IEEE.
....functional or approximate topology. The physical approach determines the physical links that connect the network together, including switches that are the internal nodes along the path. The physical approach, when combined with details about capacity and load, allows detailed modeling of LANs [2, 13] and WANs [8, 18] The functional approach makes use of end to end information. Functional topologies can be used to represent connections according to perceived performance or to represent shared media networks that are not amenable to a typical graph representation. The functional approach uses ....
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz. Topology discovery in heterogeneous IP networks. In Proceedings of INFOCOM 2000.
....software (www.peregrine.com) Riversoft s NMOS product (www.riversoft.com) and Micromuse s Netcool Precision application (www.micromuse.com) claim to support layer 2 topology discovery, but these tools are based on proprietary technology to which we do not have access. In our recent work [8], we have proposed an algorithm that relies solely on standard AFT information collected in SNMP MIBs to discover the physical topology of heterogeneous networks comprising switches and bridges organized in multiple subnets. Unfortunately, our algorithm assumes that AFT information is available ....
....the physical topology of large, heterogeneous IP networks comprising multiple subnets as well as (possibly) dumb or uncooperative elements; thus, our algorithm is essentially the first to address the the physical topology discovery problem in its full generality. Similar to our earlier work [8], the practicality of the solutions proposed in this paper stems from the fact that they rely solely on standard information routinely collected in the SNMP MIBs [5] 6] of elements and they require no modifications to the operating system software running on elements or hosts. Unlike [8] ....
[Article contains additional citation context not shown here]
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz, "Topology Discovery in Heterogeneous IP Networks", in Proceedings of IEEE INFOCOM'2000.
No context found.
Y. Breitbart, et al. Topology discovery in heterogeneous IP networks. In IEEE INFOCOM, 2000.
No context found.
BREITBART, Y., GAROFALAKIS, M. N., MARTIN, C., RASTOGI, R., SESHADRI, S., AND SILBER- SCHATZ, A. Topology discovery in heterogeneous IP networks. In INFOCOM (1) (2000), pp. 265--274.
No context found.
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri, and A. Silberschatz. Topology discovery in heterogeneous IP networks. In Proc. of IEEE INFOCOM, Tel Aviv, Israel, Mar. 2000.
No context found.
Y. Breitbart, et al. Topology discovery in heterogeneous IP networks. In IEEE INFOCOM, 2000.
No context found.
Y. Breitbart, M. Garofalakis, C. Martin, R. Rastogi, S. Seshadri and A. Silberschatz. "Topology Discovery in Heterogeneous IP Networks" In Proceedings ofIEEE INFOCOM, 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