Electrical computers and digital processing systems: multicomput – Computer network managing
Reexamination Certificate
1999-12-17
2003-03-04
Coulter, Kenneth R. (Department: 2154)
Electrical computers and digital processing systems: multicomput
Computer network managing
C707S793000, C707S793000, C707S793000, C707S793000, C709S200000
Reexamination Certificate
active
06529953
ABSTRACT:
FIELD OF INVENTION
The invention relates to distributed computer systems and more particularly to a mechanism for sharing information across the distributed system.
BACKGROUND OF THE INVENTION
Information is shared across distributed systems, or networks, generally by either a messaging scheme or a shared virtual memory mechanism. The messaging schemes typically require that the system nodes maintain up-to-date topology information. Further, the nodes must exchange certain specific series of messages to convey desired information between the nodes, and may require that the nodes send queries to determine, for example, if any changes in status have occurred. The messaging schemes are thus relatively difficult to use for certain applications. Also, if one or more of the nodes fail, the messaging schemes may become unreliable.
The shared memory systems do not scale well, since they require the nodes to coordinate their shared memory activities across the network. Accordingly, the complexity of these systems increases as the number of nodes increases.
As discussed in co-pending U.S. patent application, Ser. No. 09/335,955 entitled DISTRIBUTED PROCESSING SYSTEM WITH REPLICATED MANAGEMENT INFORMATION BASIS, which is assigned to a common assignee and is incorporated herein by reference, now U.S. Pat. No. 6,411,967, information may be efficiently provided over a distributed system using a replicated management information base (RMIB), or replicated spreadsheet, which includes columns, or categories, of information to be shared among “member” nodes and a row for each member node. Such a system scales well and uses a gossiping scheme to ensure that updated information reaches all of the member nodes. The invention discussed below is a system and a method of operating the system that improves upon the teachings of the RMIB system, to improve the scalablity of the system.
SUMMARY OF THE INVENTION
The invention is a computer network resource monitoring and location system that includes one or more hierarchical management information bases (HMIBs) through which a user can readily locate or obtain information relevant to locating and managing various network resources. The system maintains a portion of the HMIB information on each node, and a user navigates through the system starting from a local node and using contact information that is contained in the HMIBs.
More specifically, a node that belongs to a multiple-node domain, for example, a group, provides information of interest as a row of a group management information base or group MIB. The group MIB, which is in the form of a table, has columns for the different types of information of interest and includes at least (a) a node ID column for a node identifier that uniquely identifies the node that is the source of the information, and (b) a contact column for a node address and, as appropriate, other relevant contact information. Each node that supplies information to the group MIB is a member of that MIB, and each member maintains a view of the group MIB.
In systems with two or more levels, each group is a member of, for example, a multiple-group subnetwork. The system provides, to each of the nodes in the subnetwork, condensed versions of the group MIBs of each of the member groups. Each condensed version becomes a row of a subnetwork MIB, with the row including at least a group identifier that uniquely identifies the group and contact information for a subset of the nodes in the group. The condensed version may also include such information as the average or the median load on the nodes in a group, what resources are operating within the group, and so forth.
The system, in turn, provides condensed versions of the subnetwork MIBs to a next level MIB, which in the example is a multiple-subnetwork MIB. The condensed information is included as the rows of the next level MIB, with each row containing contact information for a subset of the associated nodes. The system provides condensed versions of these next level MIBs as the rows of a higher level MIB, and so forth, until the condensed versions of the relevant information are supplied to a root MIB, which in the example is a network MIB.
Each node maintains a view of each group MIB of which it is a member and also a copy of each MIB in a link from the node to the associated root, or network, MIB. Each node thus stores a portion of the HMIB information, and provides contacts through which a user can access further information about every other node in the network. For example, any node that has a copy of the network MIB can, through contact information in that MIB, direct a user to various nodes in the same or other subnetworks that have information about particular resources of interest. The user can then descend through the hierarchy of MIBs maintained by these other nodes to determine which particular nodes are operating the resources of interest. Accordingly, the local node uses the HMIB to direct the user ultimately to a node that has or is running a resource of interest, and the local node thus does not have to maintain detailed information about each of the other nodes in the network.
The system may also use a hierarchical gossiping protocol to ensure that each node receives updated information for its views of the MIBs, as discussed in more detail below. Also as discussed below, the system may employ a cryptographic scheme to prevent an interloper from corrupting the HMIB information.
REFERENCES:
patent: 5724577 (1998-03-01), Exley et al.
patent: 5864865 (1999-01-01), Lakis
patent: 5974407 (1999-10-01), Sacks
patent: 6167399 (2000-12-01), Hoang
patent: 6272537 (2001-08-01), Kekic et al.
patent: 6385604 (2002-05-01), Bakalash et al.
patent: 6411967 (2002-06-01), Van Renesse
patent: 6421658 (2002-07-01), Carey et al.
Aiken J A et al: “Achieving Interoperability with Distributed Relational Databases” IEEE Network, IEEE Inc. New York, US, vol. 5, No. 1, 1991, pp. 38-45.
Cesari and McKenna LLP
Coulter Kenneth R.
Reliable Network Solutions
LandOfFree
Scalable computer network resource monitoring and location... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Scalable computer network resource monitoring and location..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Scalable computer network resource monitoring and location... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3010338