Multiplex communications – Network configuration determination
Reexamination Certificate
1999-04-02
2001-03-20
Horabik, Michael (Department: 2735)
Multiplex communications
Network configuration determination
C370S389000, C370S401000, C709S220000, C709S223000, C709S238000
Reexamination Certificate
active
06205122
ABSTRACT:
FIELD AND BACKGROUND OF THE INVENTION
The present invention relates to a system and method for automatically detecting and analyzing the physical topology of a network, and in particular, for such a system and method in which software agents are distributed throughout the network in order to gather information about the physical topology local to each agent.
Networks, including local area networks (LAN) and wide area networks (WAN), are becoming increasingly prevalent as the number of computers in organizations grows. Networks enable information to be shared between computers, and as such are important for the case and convenience of storing and accessing data throughout an organization. Networks are implemented with a physical connection between at least two computers or other network hardware devices. This physical connection can be implemented through various types of physical connection media, including but not limited to, a cable or wireless communication, including optical, infra-red and radiowave based signals. Data is passed through this physical connection according to various protocols at different layers of the network. These protocols include but are not limited to, transmission control protocol (TCP), internet protocol (IP), internet packet exchange (IPX), systems network architecture (SNA), datagram delivery protocol (DDP) and so forth. At the data link layer, such protocols include, but are not limited to, Ethernet, token ring, fiber distributed data interface (FDDI) and so forth.
All of these different types of physical media and protocols increase the flexibility and power of network transport, thereby enabling networks to become larger and more complex. However, as networks become more complex, managing these networks becomes more difficult. Indeed, simply determining the physical topology of the network can be quite difficult. For example, if a computer having a network card with a particular MAC address is moved from one physical location to a different physical location, the physical topology of the network is changed. Such a change can have unpredictable effects for the management of traffic flow through the network. If the network administrator is not notified of this change, the source of these unpredictable effects can be difficult to locate.
In order to help the network administrator detect and manage these changes to the physical topology of the network, various solutions have been proposed in the background art. For example, U.S. Pat. No. 5,710,885 discloses a network management system for monitoring nodes in a network. This system features a central network management station which sends out polling messages to the nodes according to the IP (internet protocol) address of each node. If a node does not respond within a predetermined period of time, the node is determined to have failed. The disclosed method has the drawback of only being suitable for confirming the existence of a node at a particular IP address, such that the method would not be useful for de novo determination of the topology of a network.
Similarly, U.S. Pat. No. 5,845,081 discloses a method for detecting the existence of a network by a computer which communicates with the network through a different network protocol than that of the network itself. The computer communicates with one or more nodes in the network in order to receive information about the topology of the network. Again, the disclosed method would not be useful for de novo determination of the topology of a network.
In an attempt to provide a solution for the de novo determination of the physical topology of a network, U.S. Pat. No. 5,715,396 discloses a method for automatic topology discovery in an ATM (asynchronous transfer mode) network. According to this method, each network switch transmits messages through each port. Neighboring switches which receive these messages forward them to a topology manager which constructs a network topology profile according to the received messages. Unfortunately, this method suffers from the drawback of being restricted to switches, thereby ignoring other network components such as computers. Since a principle goal of network management is to provide more efficient traffic flow to computers on the network, such a drawback significantly reduces the usefulness of the disclosed method. Thus, although the disclosed method may be useful for ATM networks, which transmit information through the network in fixed-length cells, this method is not generally useful for the automatic discovery of the physical topology of a network.
U.S. Pat. No. 5,606,664 discloses an apparatus of limited utility for automatically determining the topology of a LAN, also described as a “concentrator” hardware device. First, the disclosed apparatus is only useful for local area networks, but cannot be employed to detect wide area networks or combinations of LAN and WAN in a network, for example. Furthermore, the apparatus is limited to the detection of network devices which are physically connected to it, which is again of restricted utility. Thus, the disclosed apparatus is not generally useful for the automatic discovery of the physical topology of a network.
U.S. Pat. No. 4,684,796 discloses an extended method for automatic topology discovery using the apparatus disclosed in U.S. Pat. No. 5,606,664, in which each “concentrator” hardware device in a network includes a network management module. This module sends a “hello data packet” to each unique network segment associated with that particular concentrator device. This data packet includes information identifying the unique segment on which the packet is being sent. The physical topology of the network is then determined according to the recipients of each packet. However, the disclosed method has the disadvantage of requiring specialized concentrator hardware devices and of requiring the basic determination of the network topology as a plurality of network segments to be performed before the topology to be determined. Thus, the disclosed method is also not generally useful for the automatic discovery of the physical topology of a network.
U.S. Pat. No. 5,708,772 discloses a method for determining the topology of a network by receiving signals from source hubs in the network, including information concerning destination hubs and the ports of the source hubs to which these destination hubs are connected. As for the methods described previously for U.S. Pat. No. 4,684,796 and U.S. Pat. No. 5,606,664, this method depends upon the use of specialized hardware apparatus. Furthermore, the method concentrates upon connections through particular network components, rather than determining the physical topology by focusing on the connections to the computers themselves. U.S. Pat. No. 5,734,824, which discloses a method for discovering network topology for local area networks connected through transparent bridges, suffers from similar drawbacks. Thus, these methods are still not suitable for general automatic discovery of physical network topology.
U.S. Pat. No. 5,850,397 discloses a method for automatic detection of the topology of a mixed-media network. According to this method, the mixed-media network is divided into groups of devices which support a common topology mechanism and which are described as “spheres”. The topology is determined for devices within each sphere, and then additional spheres are connected until the topology of the entire network is known. Unfortunately, this method has the disadvantage of requiring the division of the network into different “spheres” before the disclosed method can be performed, such that the method is not useful for de novo detection of the physical topology of a network. Thus, the disclosed method has the disadvantages of not being generally suitable for automatic, de novo detection of the physical topology of any type of network.
The recently accepted standard RFC2108 describes a system for detecting the network topology. The system features a central evaluation unit which retrieves passively collected information concerni
Cohen Yaacov
Haiut Alexander
Levy Ran
Raz David
Sharon Azulai
Friedman Mark M.
Harper Kevin C.
Horabik Michael
Mercury Interactive Corporation
LandOfFree
Automatic network topology analysis does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic network topology analysis, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic network topology analysis will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2530569