JP2002529018A - Terminal management in communication networks - Google Patents

Abstract

(57) Abstract: A connection manager for selecting some terminals from a plurality of terminals on a path that can be used in a communication network for transmitting broadband traffic. The connection manager (35) includes a purpose-specific function supported by each path of the network, a connection model (36) for displaying a position of a terminal with respect to each path, and control for managing a function specifying function of a terminal of the network. An interface (43, 44, 45, 46) to the means (81, 82, 91, 92) and a processor (41) connected to the control means through the interface and a virtual bus (42). The processor, based on the connection model (36), in response to a request for a terminal on a predetermined route of the network between the two locations, in light of the terminal function required for the predetermined route, Identify. Thereafter, the selection of the individual terminal for the path is sent to the control means identified by the processing means (41). The choice sent above will determine whether individual terminals supporting the requested function are available at the location, and the number of available terminals that support the requested function If there is more than one, including coordination between equivalent terminals.

Description

DETAILED DESCRIPTION OF THE INVENTION

FIELD OF THE INVENTION The present invention relates to connection management for large heterogeneous communication networks, such as those operated by telecommunications utilities and used by various carriers and service providers, and in particular, And a method and apparatus for selecting a terminal for connection in a broadband network.

BACKGROUND OF THE INVENTION [0002] The term "communications network" as used herein includes networks suitable for voice telephony and data communications. Such communication networks are suitable for the switched connection and transmission of voice, data, audio and / or video traffic, also referred to as broadband or "multimedia" communication.

[0003] Current communication networks are characterized by a number of transmission media using various network technologies, protocols, and software applications and devices sold by various vendors. Many devices include management functions such as monitoring, testing, and alarm functions, but centrally manage, process, and control network management functions in a complex multi-distributor environment. Is a big problem.

[0004] Customer access technologies (ADSL, HFC), core network technologies (ATM,
Another problem in heterogeneous networks, including frame relay) and transmission techniques (SONET / SDH, WDM), is that the management of terminal connections is usually based on the principle of least common divisor. The services provided by a network are limited to those services that the lowest performance device in the network can support. This principle is very inefficient in utilizing the maximum capacity of the various communication paths available in the network to meet the specific service requirements of the customer.

[0005] Conventional network management centers use a number of operational support systems (OSS) to optimize the allocation of resources for a communication network.
Normally, information on both the flow of communication traffic in the network and the functional capabilities provided by the network switching elements, collectively referred to as "network state", are stored at the network management center. Typically, information about network status is maintained in a central database that must be updated periodically to reflect changes in the communication network. The database is updated directly by agent systems located in the distributed network.
Thus, the network management center can remotely monitor network conditions to assess performance levels, detect device failures, and adjust circuit failures and network traffic.

[0006] An important issue with the central network state database is the general administrative overhead of the communication network in keeping the information up to date so that the appropriate resource allocation can be made. In some cases, this problem is also referred to as "synchronizing" the information in the network management database with the network elements, their functional attributes, and the validity of the real-time state. This problem is particularly important when managing individual terminals, which, on a large scale, affect network connectivity.

<Terms> AAD: ATM Access Device ADSL: High Performance Digital Subscriber Line ATM: Asynchronous Transfer Mode CMIP: Common Information Management Protocol CORBA: Common Object Request Broker Architecture DSL: Digital Subscriber Line EMS: Element Management system HFC: Hybrid coaxial optical fiber NMS: Network management system NTU: Network terminal unit OSS: Operation support system VPC: Virtual path connection

[0008] SDH: Synchronous digital hierarchy SNMP: Simple network management protocol SONET: Synchronous optical communication network TCP / IP: Transfer control protocol / inter-network protocol TL / 1: Network management interface protocol VCI: Virtual circuit identifier VPI: Virtual Path identifier WDM: wavelength division multiplexing.

OBJECTS OF THE INVENTION It is an object of the present invention to mitigate or overcome at least some of the problems associated with the prior art and to communicate broadband traffic between two locations in a network. To provide a connection manager for selecting a terminal for a given route.

Another object of the present invention is a method for selecting a terminal for a given path of a communication network, thereby helping to cost-effectively transfer broadband traffic between two locations in the network. It is to provide.

[0011] Other problems may be understood by reading the following description. DISCLOSURE OF THE INVENTION In one aspect, the present invention relates to a connection manager for selecting several terminals from a plurality of terminals on a path that can be used in a communication network for transmitting broadband traffic. It is. The connection manager controls: (a) a connection model that indicates the purpose-specific functions supported by each path of the network and the position of the terminal with respect to each path, and (b) means for managing the purpose identification function of the terminal of the network. (C) in response to a request for a terminal on a predetermined path of the network between two locations, and taking into account the terminal function required for the predetermined path, Processing means operative to identify the control means; (d) the selection of individual terminals for the path is transmitted to the identified control means, wherein the transmitted selection is: (i) requested Determining whether an individual terminal supporting the function can be used at the above location; and (ii) supporting and using the requested function. If the number of end there is one or more, including a decision between the equivalent of the terminal.

Preferably, the purpose-specific functions indicated by the connection model include one or more of the following: (i) a communication protocol, (ii) a transmission rate, and (iii) use of a path. (Iv) average error rate, and / or (v) physical location.

[0013] Preferably, the plurality of terminals are assigned to a terminal group according to the role of the terminal. The role of a terminal group may include location, physical configuration, or purpose-specific functions supported by the terminals in the group.

[0014] Most preferably, the terminals are managed by a hierarchy of control means. Within this hierarchy, individual terminals are located at the lowest level. If necessary, the terminals are grouped according to the hierarchy of the control means. The individual selected terminal transmitted further comprises a traversal of the hierarchy of control means up to the control means managing the individual terminal.

[0015] Preferably, the decision between the terminals is preferably made by the control means based on the cost attributed to each terminal. Preferably, the control means reports the identity of the individual terminal at each of the two locations.

Preferably, the interface to the control means is a network switching device,
Preferably, it includes a network device adapter co-located with the transmitting device and / or the terminal device.

Most preferably, when the first terminal is selected, it is preferred that the processing means operate to transmit a selection of a second terminal that can be connected to the first terminal.

Viewed from another perspective, the present invention is a method for selecting a terminal for a predetermined path in a communication network for transmitting broadband traffic. The method includes: (a) generating a connection model indicating a purpose-specific function supported by each route of the network and a position of the terminal with respect to each route; and (b) means for managing the purpose specifying function of the terminal of the network. (C) responding to a request for a terminal on a given path of the network between two locations and taking into account the terminal functions required for the given path; Identifying the appropriate control means from the connection model; and (d) sending the selection of the individual terminal for the path to the identified control means, the transmitted selections comprising: (i) requesting Determining whether an individual terminal supporting the requested function can be used in the above location, and (ii) using and supporting the requested function. If the number of terminals that can bets is more than one includes an adjustment between the equivalent terminal.

Preferably, the plurality of terminals are assigned to a terminal group according to the role of the terminal. The role of a terminal group may include location, physical configuration, or purpose-specific functions supported by the terminals in the group.

Most preferably, the terminals are managed by a hierarchy of control means. Within this hierarchy, individual terminals are located at the lowest level in the hierarchy. Preferably, the step of sending the individual selected terminals further comprises traversing the hierarchy of the control means to the control means managing the individual terminals.

Preferably, the decision between the terminals is made based on the cost of each terminal. Preferably, the method preferably further comprises the step of the control means reporting the identity of the individual terminal at each of the two locations.

The transmitted selection preferably includes the following steps, preferably in order to traverse the hierarchy of control means. The following steps are performed iteratively. (A) when the control means returns the individual terminal; (i) reporting the identity of the terminal to the connection manager; (b) otherwise, the control means comprises: (i) a lower level of the hierarchy. Sending to the control means; (ii) comparing available terminals; and (iii) returning the terminal with the lowest cost.

The step of transmitting the selection further comprises: if the first terminal has already been selected,
The method may include selecting a second terminal that can be connected to the first terminal.

Preferred embodiments will be described with reference to the accompanying drawings so that the present invention can be easily understood. DETAILED DESCRIPTION OF THE DRAWINGS Embodiments of the present invention will be described in the context of the various communications networks 10 of FIG. The connection manager of this embodiment participates in the operation process and the service assurance process of the large-scale communication network. Connection managers are ATM, SDH
It is suitable for use in connection with broadband communication products where the "network layer" (defined by the ITU-T layer management model) is quite complex, such as IP, bundled broadband products. The connection manager supports configuration and security activities at the network layer and can cooperate with other systems to perform these functions for a subset of the communication network. The connection manager of this embodiment resides in the network management layer 30 between the service layer 20 and the network element layer 40.

The service layer 20 typically creates and queries new connections, a service ordering system 21 that facilitates modification and deletion of current connections, and queries for available connection characteristics, connection costs and time frames. Includes a pre-sale system 23 that supports pre-sale activities. Examples of service layer systems include service orders, customer network management (CNM), or wholesale gateways.

Network element layer 40 is typically a network such as an ADSL / HFC customer access technology 41, an ATM core network broadband technology 42 and a transport or transmission such as a transport technology 43 such as SONET / SDH or WDM. -Includes hardware to provide services. Network element hardware can be conceptually considered to reside in different "areas",
Originally for exclusive use. Thus, the network element hardware typically uses a dedicated or compatible network element manager to represent many network elements.

Examples of network element managers include EMS systems 44 and 45 for ADSL / HFC hardware, NMS 46 for ATM core hardware,
There are vendor specific NMSs 47, 48 for the transport area. The network element manager manages many network elements, but exposes each network element as an individual component. Therefore, in other embodiments,
The connection manager can connect directly to the network element via an interface.

The network management layer 30 of this embodiment has the flexibility of a connection manager.
The first connection manager 31 manages the customer access area 40A by using the EM
Connect to S systems 44 and 45 through interfaces. The connection manager is functionally flexible because it can manage the different functional requirements of the switching matrix EMS 44 and AAD EMS 45. The second connection manager 32 manages the core area 40C, and the third connection manager 33 connects to the merchant NMS systems 47 and 48 in the transportation area 40T through an interface. The transport area has the ability of a connection manager to handle disparate merchant equipment. The connection manager uses the CMI
Includes interfaces that communicate using P, SNMP, TL / 1, or any required proprietary protocol. These interfaces can be tailored to the current or future equipment of a particular merchant.

The fourth connection manager 34 is connected to the three area connection managers 31, 32 and 33 through an interface in order to manage connections between the areas. The level of the manager or inter-region manager 34 accepts end-to-end connection instructions for all networks, determines which paths through lower networks can be used, and issues connection instructions to the region connection manager as deemed appropriate. I do.
Therefore, the connection task is sent to the appropriate domain connection manager.

Although the figure shows four individual managers, the network management layer 30
It can be considered as performing all connection management functions for the network while managing inter-region and region connections at different levels. Therefore, all network connection requirements are simplified step by step, so that each level of connection management is
Under that control, it can be optimized to manage a part of the network. However, individual connection managers can be geographically distributed through multiple sites and network operations centers, and all network connection managers 35
With a decentralized personality.

<Network Model> The connection manager is a service provider or network
Provides a flexible network modeling tool to express the ideas of the owner. The most important concepts of these expressions are as follows.

(I) A “path” that represents the owner's idea of a connection, such as ATM, PVC. (Ii) "Terminals", such as ATMs, VPIs, VCIs and cables, network ports or customer NTUs, where paths clearly appear outside the network, and the paths conceptually represent sub-paths of a sub-network.

(Iii) quality of service, bit rate or path diversity,
A characteristic that can be viewed on the terminal and selected from outside the route,
function".

Conceptually, one path can handle many network elements and protocols, such as SDH exchanges and end-to-end SDH performed by WDM transmission.

<Structure of Connection Manager> The structure of the connection manager 35 of the present embodiment, which can be developed in connection with a specific network, will be described with reference to FIG. The connection model 36 is used to display networks and their services. This model is
It can be executed by software 37. Network adapter 38 is an interface to a network element, EMS or other NMS.
The service adapter 39 is an interface for the current service OSS.

The connection manager 35 supports some basic operations related to the life cycle of a path. The network owner can order to reserve, create, or modify routes. As a result, an automatic selection, assignment and configuration of an appropriate network device can be performed to connect to a particular function between particular terminals. The network device allocated by the removal operation is released.

By using the connection manager, it is possible to determine which functions should be supported in which combinations and at what locations in the network. Terminals and routes can be searched and listed, and the terminal that best supports a given set of local features suggested by the connection manager can be determined. The create operation and the subsequent change operation make the appropriate connections.

The connection manager 35 of this embodiment preferably uses the CORBA IIOP architecture to connect through both interfaces to both the service layer and the network layer. The service layer interface and network model can be adapted to present certain standard models, such as ETSI 600-653 or ATM Forum M4, or can be adapted to current service layer interfaces. All connection manager objects can be distinguished by names and identifiers required by external systems, such as, for example, customer circuit identifiers.

The connection manager is a highly available system that supports online changes to the configuration due to backout, online database backup, copied databases, and redundant hardware. Depending on the configuration, the connection manager supports 10,000 transactions per hour on a single medium server machine (eg, a Hewlett-Packard J-Class server). This typically corresponds to a network with 50 million routes, with a normal operation latency of 0.3 seconds. Typically, the connection manager runs on a single medium-range server for one second per second.
Process a zero alarm. Usually, this processing speed is 1,000,000 to 5,0
Corresponds to setting of 00,000 routes.

As described above, one connection manager can be distributed over many server machines. Typically, the dispersion is up to 10 units. This is because, within this range, the scale of the transaction processing is almost linear. In this way, HP UX ^™ or Solaris ^™ can be successfully supported on Sun Spark ^™ , and Microsoft's NT ^™ can be supported on Intel or PA-RISC operating systems.

FIG. 3 is an overall perspective view of a theoretical connection model. The connection model is a framework for describing a communication system including a connection. More specifically, the theoretical connection model 50 of this embodiment is a distributed, object-oriented method that represents the states and operations required to manage the network layer 30 of a broadband communication network. The service layer 20 is an efficient driver for the connection model when the function supply to the service layer is the role of the connection model.

To address requests from the service layer, the connection model
3 or network element manager 54, eg, workflow manager 51,
Network manager 52, another connection manager 55 or network
Delegates to the network element layer 40 in the form of another provider at the network management layer, such as a service provider (NSP) 56. The choices involved in delegation include: (a) To which subordinate device the function is delegated? (B) How to map higher-level functions to slave devices? (C) What is the order of operation of the slave devices? (D) What action is taken if the dependent action fails? The abstract connection model 50 must be listed for operational use, along with a list according to:

(I) the particular network technology used by the network owner; (ii) the engineering rules of the network owner; and (iii) the service level requirements of the network owner.

Giving meaning to the components of the model, such as routes, functions and terminals, is a list 36 of models. Once the connection model is arranged in a list, the theoretical concept it represents has a precise meaning. In addition, the list of models has objects in the form of a list for it that indicate which objects match both the theoretical connection model and the models included in the list.

The development of a connection manager application typically involves three stages: Network analysis and design-The focus at this stage is to define the architecture of the network to be managed and analyze the characteristics of each element of the network.

Installation of a Connection Manager—The focus of this phase is to use the mechanisms supported by the core software to specify how to manage the network. The final stage in this phase is installation.

Execution Time—If a connection management system is installed, a route can be created through the network to provide communication services. <Basic Concept> The basic concept for connection management used by the connection model is the path-terminal-functional concept, as briefly described above. The function is a property of the route required by the client or customer and is specified to the route's client. Typical functions include data transfer protocol, bandwidth, reliability and error rate. For example, the ATM protocol, 64 kb / s data rate, and disability of less than 1 minute / year. Path features, such as routing through a particular network element or performing using a particular technique, are not functional. This is because the client cannot detect these characteristics. The function can be applied to a particular terminal of the route and can often be located at a connection that requires a numerical value of the function. The function of maximum bit is, for example, maximum bit rate =
Includes a numeric value that specifies the maximum bit rate, such as 256 kb / s. Functions that include numerical values that apply to the connection are called installation functions.

The route is provided by the network and is completely characterized by a route installation function (route function), a set of terminals exposed to the client, and a set of installation functions for each terminal (terminal function). The route can be permanent. That is, after the connection manager establishes a route, transmission can be performed at any time until the connection manager discards the route. The paths can be interchanged, in which case there are two stages: "configuration" and "signaling"
It is. The signaling stage starts and ends the ability to transfer data. The signaling method is issued from a network device connected to the path. The configuration phase is performed by the connection manager and establishes data transfer boundaries that can be requested by signaling. For example, the configuration allows data to be transmitted anywhere in the country's network at speeds up to 20 Mb / s. This prevents signaling that requires international transmission or 100 Mb / s transmission.

A route usually has two terminals, but the number of terminals may be one or many. Examples of routes include ATM PVC and SVC connections,
SDH connections, or customer access networks (ie, local loops). As used herein, the term path refers to the connection and trail I
It also includes the TU-T concept, and other concepts such as switched virtual connections.

Networks represent the ability to manage routes, and are used to create new routes and to list existing routes. The route is
Everything is always contained in exactly one network. A network can be conceptually viewed as a collection of pathway manufacturing plants and existing pathways. A network is also a collection of terminals whose routes are or can be specified. Exemplary networks include an ATM switch, a main distribution frame, an S
There are an ONET ring, an ATM area manager, a regional SDH network manager and the like. The term "network" includes the ITU-T concept of networks, subnetworks and network elements.

A network is typically implemented by invoking services of another network, called a subnetwork. For example, an ATM network is DS
The services of L and core ATM networks can be used. The term "subnetwork" means a client-server relationship between a network and a subnetwork. Usually, there is nothing unique about a certain network as a subnetwork. All sub-networks are completely independent networks with their own rights. Therefore, all characteristics and functions of the network are also characteristics and functions of the sub-network.

A terminal is a location where a route is, or may be, revealed to a client of the network. A terminal may also include a group of terminals.
A group of terminals may or may not be able to establish a route. Exemplary terminals include a physical port, an ATM V on the physical port.
Such as a PI or a cable pair at the customer premises. The term "terminal" includes the ITU-T concept of trail terminal points, connecting terminal points and access groups. A terminal can participate in a finite number of paths, usually one, but potentially more.

One route in the network has one or more (usually two) terminals. One terminal path can represent loopback, and multiple terminals
Multiple drops (eg, CSMA) or closed user groups (eg, voice-only networks) may be represented. Paths can share terminals, which represents multiple service capabilities (such as a set of customers using a billing server).

<Connection Manager> The connection manager 35 provides an architecture for assembling the working network management system. The core software 37 provides an abstraction of the connection model that the network owner configures to reflect the characteristics of the particular network device being deployed. Once the operational connection model 36 is configured, it substantially reflects the business and engineering policies of the network owner. That is, when it is desired to manually execute the function of the connection manager, the knowledge applied by a human operator is reflected. The core software 37
Assume that the interface to the network preferably supports a connection model 36, which is preferably represented in CORBA.

Network adapter 38 is typically a stack product such as that sold by Bartel or Hewlett-Packard to provide a simple interface to complex protocols such as CMIS or TL / 1. Developed using

The service adapter 39 provides an interface between the service management layer OSS of the network. Current operational support systems typically have dedicated interfaces, but several standards are currently being established, including the US Federal Commission's "gateway." Printed paper or character terminals are a common interface. The deployment connection manager 35 preferably has an adapter to automate the interface between the service management layer 20 and the core software 37.

<Distributed Object Model> Since the connection manager is a network layer manager, the connection manager is only relevant to network level concepts. The first network-level concept is "connection". The connection model 36 of this embodiment is:
Preferably, it is a distributed object model that is preferably represented by a CORBA interface definition word (IDL). There are three types of objects according to the concept already described. That is, (i) a route object representing a connection, (ii) a terminal object representing a location where the connection is physically clear, and (iii) a network object which is a framework capable of generating a connection.

If desired, link objects can be included to model connections between networks or sub-networks. <Network Object> The network object is a container for the route object and the terminal object. Network features form a hierarchy, where some network features are
It ranks higher with respect to other network objects. Network objects typically form a strict containment hierarchy. However, any acyclic structure can be formed by using a connection manager. The network objects can represent: That is, a group of network elements organized according to certain owner-determined criteria, such as an individual network element example, a geographic area or a functional area; managed by some other NMS, such as a merchant NMS. Subnetwork; 40A (access), 40C (ore) identified in FIG.
And an inter-area network that is a collection of several area network objects, such as 40T (transport).

The network object supports the following operations. That is, a list of capabilities of network objects, a list of characteristics of routes that the network objects can generate; a generation of routes having specific terminals and functions; a function of previewing route generation; Search for routes, terminals and subnetworks. The network object can be configured as follows. Assignment, description, semantics of identities, definition of relationships between network objects (eg, containment tree structure); definition of connections between subordinate network objects; characteristics of paths that can be created.

<Route Object> A route object represents a connection formed by a network object. The route object is
Corresponds to some real world concepts. The concepts include: (i) a physical connection, such as a bearer distribution frame; (ii) a switched connection, such as an ATM virtual circuit; or (iii) a customer and its Internet service provider (ISP). Some kind of abstract relationship, such as the relationship between

The route objects are always contained within non-network objects. If the network objects form a hierarchy, the network object can execute its path by delegating part of its execution to a sub-path in its subordinate network. A route is characterized by terminals and functions. The terminal shows where its route is revealed and the function shows features that can be seen from the outside. The route is usually
Has two terminals.

One function has a name and an optional numerical value. The function is the route itself,
Applies to terminals on the route. By doing so, terminal-specific functions can be modeled as needed for asymmetric routes. Usually, the functions interact. That is, some functions affect the ability of the path to support other functions.

Paths support life cycle type operations such as creation and deletion. With this support, several levels of path execution can be completed. The levels of execution are as follows:

(A) Design—The path does not consume any resources other than the minimum required to record its characteristics. The design state path need not follow any rules.

(B) Reservation—The route is fully executed except for the last step that causes the service to be performed. (C) Installation—A path is implemented in the device so that services can be provided.

(D) Delete-the path does not exist, but its memory is maintained for auditing purposes. A path has a cost that represents the amount of resources required to execute the path. This cost allows the client to make a reasonable choice between several candidate paths, each of which can support its needs.
The route object supports the following operations: delete, change function or terminal, preview these operations, and creates a list of route attributes.

<Link Object> A link object is used to model a connection between networks. Link objects typically represent another lower layer of the network protocol stack. For example, the link of the IP network may be an ATM. On the other hand, the link of the ATM network may be SDH. A link has a terminal group at either end on some subnetwork. A link may also specify one or more route objects that implement the link. For example, an ATM network may specify a link between two regional ATM sub-networks running using a path from the SDH network.

Link objects have a certain cost and can model the available and used capacity of the interconnect they represent. The link object supports the following operations: change execution path, delete link, describe link, and indicate the "used" capacity and "usable" capacity of the link.

<Terminal Object> For the terminal object, the route object is specified (or may be specified).
Indicates the location. The terminal objects correspond to, for example, some real world concepts. The concept is: a physical terminal, such as a cable; a single channel multiplexed on some kind of bearer, such as an ATM virtual circuit or SDH container;
Grouping of multiplexed channels, such as ATM virtual paths; one terminal object within one network object, etc. The network can effectively represent an infinite number of terminals. For example, an ATM network can model each VPI / VCI as a terminal. The rougher modeling than the ATM example,
Has many terminals. In order to be able to manage them easily, the connection manager usually supports the grouping of terminals into "terminal groups".

<Terminal Group> The connection model preferably manages the commonality of terminals by using terminal groups. In the case of this embodiment, the terminal group of one network object preferably forms an accommodation tree 60 as shown in FIG. The tree contains a root node 61 containing all terminal groups, but without forming the tree, these terminal groups remain unaccommodated. The upper layer of trees is usually an abstract group based on physical location, usually a city 62 or central office building and switching equipment 63. The lower layer of the tree is typically more specific, such as an interface card 64 or cable 65. Tree leaves are the lowest level of a modeled terminal such as a cable, IP address 66, virtual channel 67, or similar technical structure.

The use of terminal groups for devices such as exchange N2 is as follows. -All ATM VCIs are terminals in VPI terminal group 2. -VPIs in the same cable are included in the cable 2 group. Cables ending with a specific card are grouped on card 2 Finally, it is placed in the group for switch stage N2. The connection manager does not clearly distinguish between terminals and terminal groups-terminal groups are located at higher levels in the tree and individual terminals are located in the direction of the leaves. Network objects typically indicate these levels in the containing tree that are preparing to establish connections between those that can be restricted to leaf nodes. The meaning of the terminal and the structure of the terminal group are determined by the configuration of the core software.

[0072] Preferably, each terminal preferably displays to the client the cost of supporting a path including a specific function. By doing so, each can make a reasonable choice among several possible terminals that can meet its needs. If cost is not an issue, it may be desirable to make coordination between equivalent terminals by making a random selection or perhaps based on a physical sequence within the associated network device.

The terminal object has the following operations: terminal description; terminal group description; navigation within the terminal group structure; function-specific set within the terminal group; It can support finding the cheapest available terminal. A typical network has a very large number of terminals,
Most deployments configure terminals at the level of the terminal group. The characteristics of the individual terminals come from the network equipment adapter. The terminal group object supports the following configuration. That is, it supports the terminal group structure and (if desired) the cost of group members.

<Role of Terminal> Each terminal (and any corresponding terminal group) has a purpose called “terminal role”. An example of a terminal role 70 is shown in FIG.
2, a card 73, a cable 74, an IP address 75, an ATM VPI 76 and an ATM VCI 77. Since the components can reflect certain dynamic characteristics of the network, there is no specific requirement that the terminal group must have a countable or a fixed number of components. For the normal hierarchy of terminal groups, the role of the terminals tends to reflect the position of the higher layers and tends to merge with lower level functions. Referring to FIG. 4, a request for an available terminal in the terminal group "Sydney" including the role "ATM VPI" for the connection merging device can be returned, for example, as follows.

[Sydney. North. SwitchN2. Card2. Cable2]
. 1) The connection manager can use any level of terminal grouping required by the service provider or network owner. Using the above example, the connection manager can assign a route at the interface card level 64 and send the interconnect to the cable level 65, while the availability of the function is at the switch level 63 Can be processed. Each level can be mixed. For example, costs can be specified at the card and cable levels. In some cases, a terminal group may correspond to a customer network terminal unit (NTU). It is important that the scheme for the terminal group selected by the network owner or telecommunications engineer is clear for any individual terminal.

Network Interface The connection manager can interact with the devices of the network in a number of different modes. The connection manager is typically an SNMP, TL / 1 or CM
Through the IS system, it can be directly interfaced with network elements. The connection manager can interface with the vendor's element manager, in which case the connection manager checks the individual network elements. The connection manager can connect to the vendor's area manager by an interface, in which case the area manager is used to establish a sub-path within that area, or the connection manager is connected to another network. -Can be connected to the owner's wholesale sales network via an interface. The connection manager uses the public CORBA IIOP interface to the network to display the life cycle, model, state, search and sub-functional operations described on page 10. In order to deploy a connection manager, it is usually necessary to supply a network device adapter from the IIOP interface to any interface presented by the network layer.

EXAMPLE FIG. 4 illustrates the selection of a terminal according to the preferred system and method of the present invention.
5 together with a drawing showing the terminal groups of FIG. FIG. 5 shows the equipment associated with switch N2 and switch E2 including their respective adapters in north Sydney and east Melbourne, along with interconnection to a connection manager located at the telecommunications company's national network management center. . In this example, the client has a bit rate of 1 between North Sydney and East Melbourne.
. ATM with a function of 5 Mbps, which must be used for more than 4 minutes per year
Requesting a virtual channel, so that the connection manager from the northern Sydney subnetwork (N) via the central Sydney subnetwork (C) and also via the Canberra network and the East Melbourne subnetwork (E) Assign the route connected to.

For simplicity, the connection manager 35 is shown as a single component, but is typically located on a number of the above computer sites (such as the Sydney and Melbourne capital administration sites). In addition, it should be understood that they are usually implemented in a distributed manner. Including terminal functions to support ATM virtual channels,
In response to a request for a terminal on the route between North Sydney and East Melbourne, the processor 41 determines from the connection model 36 a terminal group North corresponding to the North Sydney switch 80 and a terminal connection East corresponding to the East Melbourne switch 90. Identify.

According to a preferred method for selecting a terminal, the connection manager 35 communicates with the exchange manager 81 to the CORBA standard virtual bus 42 with the network adapter 44 which is connected through an interface. Send the appropriate terminal selected in Sydney to the exchange manager 81. The switching manager 81, which controls the operation of each exchange N1, N2 and N3, determines that the terminal group N2 associated with each exchange supports the purpose-specific functions required for the ATM channel terminals. Therefore, the selection of the terminal from the terminal group N2 is sent by the exchange manager 81 to the ATM port manager 82 through the second network adapter 43.

The ATM port manager 82 controls the interface card 8 for the exchange N2.
3 controls the allocation of ATM ports that can be used on the cable attached to Therefore, the port manager is associated with a terminal role card 73, a cable 74 and a terminal group having an ATM virtual port (VPI) 76 and a channel (VCI) 77. The port manager 82 has the lowest cost "
Leaf "or an individual terminal, that is," Sydney. North. N2.C
ard2. Search within the terminal group until the ATM virtual channel identified as "Cable 2.2.1" is returned to the connection manager. Preferably, the recursive search algorithm used in the preferred method can be represented by the following pseudo code.

[0081]

[Table 1] The connection manager also communicates the exchange manager to the CORBA standard virtual bus 42 with the network adapter 45 connected through the interface to initially select the terminal in East Melbourne.
Send to The exchange manager 91, which controls the operation of each of the exchanges E1 and E2, determines that the terminal groups E1 and E2 associated with each exchange have the purpose-specific functions required for the ATM channel terminals. However, the cost of E1 is 1.5 Mb
It is much cheaper for certain features at the ps bit rate. Therefore, the selection of a terminal from the terminal group E1 is made through the second network adapter 46 through the ATM
Sent to port manager 92.

This transmission process is repeated by the connection manager 35 for each intermediate sub-path through the central Sydney network and the Canberra network, if the sub-path requires a terminal.

FIG. 6 shows how to execute the method of the present invention in a sub-network 100 that includes a partial usage exchange connection. Supports transmission rates of 1.2 Mbps, requirements 102 for routes to a specific terminal T _x in Melbourne from the appropriate terminal in Sydney is received by the connection manager 101. The connection manager 101 is related to the access area manager 103, and the access area manager 103
A series of DSL multiplexers A ₁ , A ₂ ,. . . , In charge of the A _n. The connection manager 101 further sets A
Associated with the TM area manager 105, this ATM area manager 105 is likewise a set of ATM switches S ₁ , schematically represented as ATM connection models 106.
, S ₂ , S ₃ and S ₄ .

The requirement 102 is processed by the connection manager 101 and has a function of 1.2 Mbps.
Is sent to the first control means in the form of the access area manager 103 as a request 107 for an input terminal in Sydney that supports The area manager determines that any of the input terminals T ₁₁ -T _nm for multiplexer A is 1.2 Mbp
s is supported, and T ₁₄ is randomly selected from those terminals currently available. Thereafter, the identification of the input terminal T _14, access
The message is returned 108 to the connection manager 101 by the manager 103. Thereafter, the connection manager, the request for the output terminal that can be connected to T ₁₄ 109
The feed to the access manager 103, the access manager 103 sends back 110 to the output terminal T _06.

The connection manager 101 uses the DSL multiplexer managed by the access manager 103 based on the connection model 112 to
To supply a terminal for the link 113 between the A exchange and the ATM exchange S, which is managed by T., and that T ₁₆ is the corresponding terminal of the terminal T ₀₆ . Connection manager 101, route connecting the T ₁₄ to T _06, and T ₁₆ to monitor the installation of a path connected to the T _x.

Overview The connection manager of the present invention operates at several levels. The inter-region management level can receive end-to-end connection commands over the entire network, route the underlying sub-networks available for it, and send connection tasks to multiple network region management layers. The network area connection manager recognizes the terminal role of the related terminal group, but sends the selection of each terminal to the subordinate control system managing the function specifying function. The connection manager does not need to track the availability of network elements and the real-time status and capabilities of the associated individual terminals, thus reducing the flow of management information.

By automating the transfer and configuration of connections through complex networks, connection managers significantly reduce the need for manual management at the network and element levels. The real-time state of the network is dynamically determined as needed, reducing the hassle requirement for synchronization with a central network management database. Connections can be made in real-time, and connection managers grow as broadband communication networks evolve, and as they grow, to handle new connections.

A method for performing object-oriented modeling in a connection manager. in this case,
The abstract connection model is specialized from the connection model list, so that clients and servers can be highly reused. It also allows you to match rapidly changing business scenarios,
It can run very flexible clients and servers, but does not enforce it.

While the preferred embodiment of the invention has been described throughout the specification, the invention is not limited to any embodiment or particular combination of features.

[Brief description of the drawings]

FIG. 1 illustrates various different communication networks including a hierarchy of connection managers.

FIG. 2 is a view showing the structure of a connection manager according to the first embodiment;

FIG. 3 is a diagram showing an overall situation from a perspective view of an abstract connection model of the first embodiment.

FIG. 4 shows a hierarchical tree structure of a group of terminals that can be located in a communication network.

FIG. 5 shows a portion of the network of FIG. 4 including an expanded connection manager that enables selection of transmitted terminals.

FIG. 6 shows a preferred method for transmitting a selection of terminals of a network using a partial utilization exchange connection.

────────────────────────────────────────────────── ─── [Continued from summary] Includes coordination between equivalent terminals when there is more than one available terminal in use.

Claims (20)

[Claims] 1. A connection manager for selecting some terminals from a plurality of terminals for a path that can be used in a communication network for transmitting broadband traffic, the connection manager comprising: (a) a network; (B) a connection model that displays a function for each purpose supported by each route and a position of the terminal with respect to each route; and (b) an interface for managing and controlling the purpose specifying function of the terminal of the network; c) operable in response to a request for a terminal on a predetermined path of the network between two locations, in light of the terminal functions required for said predetermined path, to identify appropriate control means from the connection model. Processing means, and (d) the selection of individual terminals for said path is transmitted to the identified control means. The transmission selection comprises: (i) determining whether an individual terminal supporting the requested function is available at the location; (ii) using, supporting the requested function. Connection manager, including coordination between equivalent terminals when there is more than one possible terminal. 2. The connection manager according to claim 1, wherein the purpose-specific function represented by the connection model includes one or more of the following: (i) a communication protocol; and (ii) transmission. Speed, (iii) path utilization, (iv) average error rate, (v) physical location. 3. The connection manager according to claim 1, wherein the plurality of terminals are assigned to a terminal group according to the role of the terminal. 4. The connection manager according to claim 3, wherein said roles of a terminal group include location, physical configuration, and purpose-specific functions supported by the terminals in said group. 5. The connection manager according to claim 1, wherein the terminals are managed by a hierarchy of control means, wherein the individual terminals are located at the lowest level in the hierarchy. Connection manager. 6. The connection manager according to claim 5, wherein said terminals are grouped by said hierarchy of control means. 7. The connection manager according to claim 5, wherein the selected individual terminal transmitted further comprises a traversal of the hierarchy of control means towards a control means for managing the individual terminal. Connection manager. 8. The connection manager according to claim 1, wherein adjustment between terminals is performed by said control means based on a cost for each terminal. 9. A connection manager according to claim 1, wherein said control means reports the identity of said individual terminal at each of two locations. 10. The connection manager according to claim 1, wherein the interface to the control means includes a network exchange, a transmission or a network equipment adapter co-located with a terminal equipment. manager. 11. The connection manager according to claim 1, wherein the processing unit can connect to the first terminal when the first terminal has been selected. Connection manager to transmit the selected second terminal. 12. A method for selecting a terminal for a predetermined path of a communication network for transmitting broadband traffic, comprising: (a) a purpose-specific function supported by each path of the network; Generating a connection model representing a location; (b) providing an interface to control means for managing the purpose-specific function of the terminal of the network; and (c) providing an interface to the network between two locations. Identifying a suitable control means from the connection model in response to a request for a terminal on the predetermined route in light of a terminal function requested for the predetermined route; and (d) individual terminals for the route. Transmitting said selection to said identified control means, said transmitted selection comprising: Determining whether an individual terminal supporting the function is available at the location; and (ii) determining if there is more than one available terminal supporting the requested function. And coordination between equivalent terminals. 13. The terminal selection method according to claim 12, wherein the plurality of terminals are assigned to a terminal group according to the role of the terminal. 14. The terminal selection method according to claim 13, wherein the role of a certain terminal group includes a position, a physical configuration, and a purpose-specific function supported by terminals in the group. 15. The terminal selection method according to claim 12, wherein the method further comprises a step of the control unit reporting the selection of the individual terminal. 16. The terminal selection method according to claim 12, wherein the terminals are managed by a hierarchy of control means, and in this case, the individual terminals are set to a lowest level in the hierarchy. How to select the terminal that is located. 17. The terminal selection method according to claim 16, wherein transmitting the selection of the individual terminal further comprises traversing the hierarchy of control means toward control means for managing the individual terminal. Selection method. 18. The terminal selection method according to claim 16 or 17, wherein the transmitted selection comprises the following steps, which are performed recursively, preferably to traverse a hierarchy of control means. Selection method: (a) when the control means returns the individual terminal; (i) reporting the identity of the terminal to the connection manager; (b) otherwise, the control means: Transmitting to the lower level control means in the hierarchy; (ii) comparing available terminals; and (iii) returning the terminal with the lowest cost. 19. The terminal selection method according to claim 12, wherein adjustment between terminals is performed based on a cost of each terminal. 20. The terminal selection method according to any one of claims 12 to 19, wherein the step of transmitting a selection further comprises the step of: A terminal selection method including a step of selecting a second terminal that can be connected.