This is a U.S. national stage of PCT application No. PCT/EP99/10363, filed on Dec. 23, 1999. Priority is claimed on that application.
The present invention relates to distributed ATM (Asynchronous Transfer Mode) connection control, and in particular to a method of managing connections in an ATM network element with distributed connection control comprising a plurality of control units, an ATM network element with distributed connection control comprising a plurality of control units and a telecommunications system comprising at least one ATM network element according to the present invention.
In an ATM network element acting as a switch in an ATM based telecommunications network, connection control is distributed to plural control units. The distributed ATM connection control poses a problem due to the fact that information distributed within the connection control has to be consistent and correct throughout the network element. In this context, distributed information refers to basic ATM logical resources such as interface, virtual path termination point (VPtp) and virtual channel termination point (VCtp) and the connections between these instances.
Because of the distributed connection control it is not transparent how a connection is created between termination points. Thus, it cannot be ensured that a connection affects only one unit of connection control and that the termination points connected are controlled by the same control unit.
It is therefore an object of the present invention to provide distributed control management in an ATM network element comprising a plurality of control units.
According to a first embodiment of the present invention, this object is achieved by a method of managing connections in an ATM network element with distributed connection control comprising a plurality of control units. According to
Moreover, according to the first embodiment of the present invention, the above-mentioned object is solved by an ATM network element with distributed connection control as shown in
According to the first embodiment, connection control is divided into a plurality of management areas each being under control of one control unit. This control unit is in charge of all interfaces inside the respective management area, an interface comprising at least one termination point. The control unit knows all logical resources inside its management area.
The managing unit stores interface identifiers and determines the control unit in charge using the interface identifiers. The managing unit can be implemented in the network element as file, library or program.
In a telecommunications system comprising at least one ATM network element according to the first embodiment of the present invention, in case of a failure of a control unit, termination points and connections within the management area being controlled by the control unit are lost. However, since a connection is handled by two control units, it can be saved. This requires that each control unit gets notification of a failure of the other control units. Furthermore, it is required that the control units get notification about restart of the other control units. After receiving a control unit restart notification, each control unit searches all termination points in the connections related to the restarted control units. Then, all connections to the restarted control units are re-established by sending re-establish requests.
If one control unit would handle termination points of only one logical interface, all connections could be saved in the case of a failure. Anyhow, this would not be appropriate since there are too many logical interfaces compared with the number of control units where the connections could be handled.
In order to solve such problem, a spare connection control can be provided. With the spare connection control, a spare connection control unit for connections inside one management area or control unit area is provided. This spare control unit actually is the (main) control unit for some other area. Now, when the managing unit receives a connection request that would be handled inside one connection control unit (i.e. one management area) for both termination points, it also determines a spare control unit in which the connection information is stored (i.e. the connection information is duplicated). Now, the connection information is present in both the spare control unit and the (main) control unit. In this case, if the (main) control unit suffers a failure, the original situation can be restored by loading data from the spare control unit. Thus, all connections can be saved in the case of a failure.
With the first embodiment of the present invention it can be ensured that information distributed in the connection control is consistent and correct throughout the network element.
The distribution of connection control functionality according to the first embodiment of the present invention suggests that the connection admission control functionality should follow the same distribution method.
According to a second embodiment of the present invention, the above-mentioned object is achieved by a method of managing connections in an ATM network element with distributed connection control comprising a plurality of control units. As shown in
Moreover, according to the second embodiment of the present invention, the above-mentioned object is achieved by an ATM network element with distributed connection control as shown in
According to the second embodiment of the present invention, the determination of the control units is performed by the managing unit on the basis of a selected algorithm. This algorithm may be a consecutive selection of one unit out of existing working units. A further selection method may also take into account the amount of maintained instances of termination points in the handling units.
The first and second control units handle the first and second termination points, respectively, and the third control unit handles the requested connection. The managing unit stores information about the termination points and the requested connection. The determined control units control the sequence needed for topology requests, acceptance from connection admission control and information delivery to ATM layer functions.
In a telecommunications system comprising at least one ATM network element according to the second embodiment of the present invention, ATM connection control functionality and data storage is distributed to several parallel units. One control unit (control unit 3 in
In the following the present invention will be described by way of preferred embodiments thereof with reference to the accompanying drawings 5 to 8.
In the following, a first embodiment of the present invention will be described by referring to
The first embodiment provides a solution of the object of the present invention, which is based on the topology of the ATM network element. In the network element, the locations of different logical interfaces are distributed. According to the first embodiment, interfaces within one management area in the network element are assigned to one connection control unit. The connection control unit is responsible for all logical resources inside this management area, the logical resources comprising interfaces, VPtps (Virtual Path termination points) and VCtps (Virtual Channel termination points).
In
As can be seen from
In case of a connection between interfaces (or VPtps, OR VCtps), connection information is distributed with the termination point information of the interfaces to be connected. In other words, when a connection between termination points of two different interfaces is created, information about the termination points to be connected is stored at two different places. According to the first embodiment, the control units in charge of the termination points to be connected store both information about their own termination point and information about the other termination point being part of the requested connection.
Service providers are located in the ATM telecommunications system by means of management units or brokers. A broker is implemented in the systems such that for a given logical interface the broker determines one connection control unit to be used for this interface. The broker service can be provided in the system by means of a file, a library or a small program managing the required data structure.
The broker is used by users of the connection management service. User requests are typically create, delete or modify requests for termination points and connections between termination points of different interfaces. Common to all these requests is that they relate to one interface (one termination point) or plural interfaces (plural termination points). Therefore, the interface identification is very suitable. For this purpose, the broker stores interface identifiers indicating the management areas to which the interfaces belong in order to determine the control units in charge of the interfaces.
According to the first embodiment, the connection control units use the broker in order to find out the other connection control units being part of a request connection between termination points. A connection between termination points is handled by storing in the control unit being part of the connection the “own” termination point information and the information of the termination point belonging to the other control unit. This leads to a negotiation between connection control units in which one control unit exchanges the termination point information with an other control unit holding the information of the other termination point in the connection.
Then, in communications 6, the control units 1 and 2 request the respectively other control unit forming part of the connection between the termination points. In other words, connection between the termination points is requested by the control units 1 and 2, and in communications 7, a connection check of the requested connection is performed. Finally, in communications 8, the control units 1 and 2 each perform hardware update.
When in the case of a failure termination points and connections of an interface controlled by a control unit are lost, according to the first embodiment, the connection handled by two control units can be saved. This requires that each connection control unit is notified by the other control unit of a failure. Furthermore, it is required that the control units are notified of a restart of other control units. After receiving a connection control restart notification, each control unit tries to find out all the termination points in the connections related to the restarted control units. All the connections with the restarted control units are then re-established by sending re-establish requests.
If one control unit would handle termination points of only one logical interface, all connections could be saved in the case of a failure. However, this would not be feasible because there are too many logical interfaces compared with the number of control units handling the connections.
In the following, a second embodiment of the present invention will be described by referring to
The broker selects termination point identifier values, i.e. the free VPI (virtual path identifier) and VCI (virtual channel identifier) values, for a requested termination point of an interface. The identifiers are stored in a connection information table (conn info table) of the broker. Data or attributes of the respective termination point identifiers (TPIs) are stored in termination point information tables (TP-info tables) of the respective control units. The selection of the handling control unit is performed by the broker by means of a selected algorithm. For example, this algorithm can be a consecutive selection of one unit out of existing working units. Moreover, a selection method can also take into account the amount of maintained instances of TPs in the handling units. The broker also performs conformance definition checking for the TP and connection requests.
In the distributed parallel control units, the specific termination point and connection requests are handled. For each request a hand is created in the respective control unit to handle the tasks needed for the request. The hand is only alive the time needed for the request.
Each control unit stores the termination point information in its TP-info table and handles the sequences needed for topology requests, acceptance from connection admission control (CAC) and information delivery to ATM layer functions.
According to
In the control units 1 and 2, respectively, a hand is created to perform the requested tasks. The termination point information about the first termination point is stored in the TP-info table of the control unit 1 and the termination point information about the second termination point is stored in the TP-info table of the control unit 2. The sequences required for topology requests, acceptance from CAC and ATM layer services are handled by the control units 1 and 2. Thereafter, each unit hand returns a record index of the termination point information record stored in the respective TP-info tables to the broker hand. The broker updates the received termination point record index and information about the handling control units (units 1 and 2) in the termination point information record in its connection information table.
Then, the broker determines a control unit to handle the requested connection. As shown in
According to the second embodiment of the present invention, ATM connection control functionality and data storage is distributed to several parallel units. One control unit (control unit 3) is aware of the connection between two termination points each handled by different control units (control units 1 and 2). Thus, it can be ensured that the termination points connected are under the same unit of connection control.
While the invention has been described with reference to preferred embodiments, the description is illustrative of the invention and is not to be construed as limiting the invention. Various modifications and applications may occur to those skilled in the art without departing from the true spirit and scope of the invention as defined by the appended claims.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/EP99/10363 | 12/23/1999 | WO | 00 | 9/19/2002 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO01/49063 | 7/3/2001 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
5838921 | Speeter | Nov 1998 | A |
6343322 | Nagami et al. | Jan 2002 | B2 |
6606310 | Vialen et al. | Aug 2003 | B1 |
6714989 | Mizutani et al. | Mar 2004 | B1 |
6781994 | Nogami et al. | Aug 2004 | B1 |
Number | Date | Country |
---|---|---|
WO 9515635 | Jun 1995 | WO |