Communication system and apparatus

This application is a continuation application of International PCT Application No. PCT/JP2007/000322, filed on Mar. 28, 2007.

The embodiments discussed herein are related to technology which provides a telecommunication service using MPLS (Multi Protocol Label Switching), and are specifically related to a communication system which provides redundancy.

Recently, MPLS has received a lot of attention as packet transmission technology. The IETF (Internet Engineering Task Force) is promoting specification study and standardization of PWE3 (Pseudo Wire Emulation Edge-to-Edge), which provides existing services (such as FR (Frame Relay), ATM, TDM, and Ethernet) on an end-to-end (point-to-point) basis utilizing MPLS as tunneling technology. From the standpoint of service integration, Pseudo Wire (PW) technology utilizing MPLS is expected to expand into carrier networks and will require further higher reliability in the future.

FIG. 1 illustrates a model diagram of PWE3 defined by IETF (RFC (Request For Comments) 3915, 3986). FIG. 1 illustrates a PSN (Packet Switched Network) 1 in which a PSN tunnel 2 and Pseudo Wire PW are established between provider edges PE 1 and PE 2 . In the Pseudo Wire PW, PW 1 and PW 2 are established in correspondence with customer services, for example. The provider edge PE 1 is connected to a customer edge CE 1 outside the Pseudo Wire PW via an attachment circuit 3 , and the provider edge PE 2 is similarly connected to a customer edge CE 2 outside the Pseudo Wire PW via an attachment circuit 4 .

Here, an end-to-end service 5 (such as Ethernet, ATM, Frame Relay, TDM, and PPP/HDLC) is emulated using a packet network. As technology which provides virtual pipes (tunnels), MPLS and L2TP (Layer 2 Tunneling Protocol) are mainly used.

In carrier networks, high reliability must be ensured. As an example, it is important to construct efficiency-focused connection management technology, and a mechanism in which for example a single failure in a network does not become a load on end users is required.

ITU-T Recommendation Y.1731, which specifies carrier Ethernet OAM (Operation, Administration, and Maintenance) has realized a mechanism considering the above. The specific configuration is illustrated in FIGS. 2A , 2 B, and 2 C. In the Ethernet service illustrated in FIGS. 2A and 2B (i.e., a service to aggregate traffic between CE 1 -[1, 2, . . . , n] CE 2 -[1, 2, . . . , n] at PE 1 and PE 2 ), Y.1731 specifies that a provider layer OAM management domain be provided in a provider domain PE 1 -PE 2 , and also that a customer layer OAM management domain be provided in a CE domain, and this customer domain is set in a manner that overlays the provider domain. FIG. 2B illustrates a case in which this system has a hierarchical structure including a customer layer 21 and a provider layer 22 .

In this case, alarm forwarding when a single failure occurs in the provider management domain 22 between PE 1 -PE 2 is performed in a manner such that an apparatus at the failure end generates an AIS (Alarm Indication Signal), detects LOC (Loss of Continuity), and notifies PE 1 and PE 2 of the AIS. Receiving this notice triggers PE 1 and PE 2 to generate an AIS notice for the customer domain and deliver it to the end point. On receiving the AIS, an alarm inhibit function operates to inhibit LOC detection, and at the same time a failure in the service domain may be detected and recognized.

FIG. 2C illustrates a case in which there is no hierarchical structure. While a failure in the service domain may be detected by not receiving a CC (Connectivity Check) transmitted from the customer device, receiving no CC may be detected in principle with a time-out of a timer having a length several times longer than a transmission interval. Therefore the detection time is estimated to be much longer than that in FIG. 2B . While reducing the detection time may be achieved by speeding up (shortening) the CC cycle, in that case, not only might bandwidth in the provider be suppressed, but also processing at PEs may be burdened with a load.

As described above, layer management in a packet network is important to achieving flexible connection management for each customer. Considering the above, it is necessary to construct layers such those as illustrated in FIG. 3 in PW/MPLS also, and based on that, a mechanism for providing OAM, connection management, and failure notification is required. Specifically, the service layer 31 corresponds to the customer layer 21 , and the Pseudo Wire (PW) layer 32 and the MPLS layers 33 and 34 correspond to the provider layer 22 . Packets consisting of L 2 and OAM (FDI, etc.), and packets consisting of L 1 , P 1 , and OAM (BDI, etc.) are bidirectionally transmitted from PE 2 to PE 1 , and from PE 1 to PE 2 , respectively. When a failure is reported/detected in the MPLS layer 34 , escalation to the PW layer 32 and the service layer 31 is required. In other words, since each layer is independent of the others, it is necessary to perform LOC detection by timing out for each PW unit at PE 1 . Even in the case of a failure in the MPLS LSP (Label Switching Path), there is no mechanism to notify the PW layer 32 of an alarm, and as a result, a failure is detected by Loss Detection (timer-dependent) which may be not only time-consuming, but also complicated since one-to-one processing is performed for each PW.

In addition, it is necessary to insert flags equivalent to RDI (Remote Defect Indication) on the opposite part. Unlike what is illustrated in FIGS. 2B and 2C , the MPLS layers 33 and 34 existing under the PW layer 32 have two unidirectional paths. Therefore it is necessary to assume a unidirectional failure and thus that RDI is a necessary function. Consequently, this optional function to insert the RDI is further required. In the MPLS layers 33 and 34 , communication is performed through tunnels 35 and 36 .

Considering the above, processing for each PW becomes complicated as in FIG. 2B in the PW/MPLS above. And as was described in the beginning, it is desirable that PE 1 be notified of a failure on the basis of a single failure occurring below the MPLS layer, and then RDI may be generated. However, this has not been achieved satisfactorily with the current framework. Specifically, escalation to the PW layer 32 and the service layer 31 cannot be achieved when a failure is notified/detected in the MPLS layers 33 and 34 .

This situation is illustrated more specifically in FIG. 4 . OAM, which is equivalent to AIS occurring in a lower layer of PW (ITU-T Y.1711 FDI, for example), is received at PE 1 . PE 1 needs to extract labels P 1 - 1 ˜P 1 - n to identify opposite customer services with L 2 or LSP 2 (MPLS label) to identify this AIS equivalent OAM.

However, the current framework of signaling (label distribution) is independent for each layer (MPLS, PW). Practically, RSVP-TE defined in RFC3209 or LDP (Label Distribution Protocol) defined in RFC3036 are used in MPLS, and LDP defined in RFC4447 (RFC3036) is used in PW. These are performed in independent procedures based on IP (Internet Protocol). In other words, there is a possibility that a signal for signaling or a label mapping message may not be transmitted on the MPLS tunnel in a configuration of PW, and thus the opposite labels of the tunnel-directed label L 2 or the customer service identification labels P 1 - 1 ˜P 1 - n may not be extracted.

While patent document 1 discloses alarm forwarding in a network, it is different from embodiments of the present invention in that a wavelength is used in a unidirectional link.

Patent Document 1: Japanese Laid-open Patent Publication No. 2004-312152

An embodiment of the present invention provides a management system of labels and paths applied to a communication apparatus using PW (Pseudo Wire) technology on the basis of MPLS. It allows an LOC function from the lower layers (below MPLS) to be reported (escalated) to PW by providing a function having a correlation between MPLS (layers) and PW (layers), and is characterized by its efficient failure notification. Practically, such function is achieved by extending PW signaling.

The first aspect of the present invention provides a communication system which is a transmission system using MPLS (Multi Protocol Label Switching) tunneling and label technology to map a communication service to the MPLS and to provide bidirectional services on an end-to-end basis. The communication system is characterized by including a means to perform label assignment control for the bidirectional services; a means to transmit, after completion of the assignment control, a message in which an LDP message used in the assignment control is a payload and a tunnel-directed label (L 2 ) used by the bidirectional services is a header; and a means to associate, on the basis of the received LDP message used in the assignment control, a service-directed label (P 1 ) included in the corresponding LDP message with the received tunnel-directed label (L 2 ), in the part where the message is terminated.

This allows the customer service label P 1 from the first edge to the second edge to be associated with the customer service label P 2 from the second edge to the first edge in MPLS bidirectional communication system. Therefore, the tunnel label L 2 in a certain direction in the MPLS layer may be escalated to the customer service identification label P 2 in the certain direction in the PW service layer, to the customer service label P 1 in the opposite direction, and to the tunnel label L 1 in the opposite direction.

The second aspect of the present invention provides a communication system to perform the bidirectional communication services of the first aspect, and is characterized in that the means to associate includes a means to analyze the message to determine the presence of a message having the same Forwarding Equivalence Class (FEC) or the same forwarding attribute in the label assignment; and a means to look up labels (P 1 - 1 ˜P 1 - n ) transmitted from the first end to the second end which identify the customer service, and to detect corresponding labels (P 2 - 1 ˜P 2 - n ) transmitted from the second end to the first end which identify the customer service.

The third aspect of the present invention provides a communication system according to the first aspect of the present invention, and is characterized in that the header of the message includes a specific label for control management.

The fourth aspect of the present invention provides a communication system according to the first aspect of the present invention, and is characterized in that the header of the message includes a specific data field for control management.

The fifth aspect of the present invention provides a communication system according to the first aspect of the present invention, and is characterized in that the header of the message includes IP and UDP.

The sixth aspect of the present invention provides a communication system according to any of the first to fifth aspects of the present invention, and is characterized in that, when receiving a failure notification message in which a tunnel-directed label used by a service is a header, a service-directed label in the opposite direction is obtained, and a payload associated with the obtained service-directed label reflects that a service using a tunnel to which the received failure notification message belongs is in a disconnection state.

This allows correspondence between labels L 2 and P 1 - n , for example to transmit data including the label L 2 and OAM (FDI, etc.) from the provider edge PE 2 to the provider edge PE 1 , and to transmit data including the label L 1 , label P 1 - n , and OAM (RDI, etc.) from the edge PE 1 to the second edge PE 2 for failure notification. In other words, if a failure occurs in a path toward one provider edge PE 1 , a packet having the label L 2 and the customer service identification label P 2 as management codes followed by management data is transmitted to notify PE 1 of the failure. Then PE 1 may transmit data having the label L 1 and the customer service identification label P 1 as management information to the other provider edge PE 2 through a path without a failure.

The seventh aspect of the present invention provides a communication system according to any of the first to fifth aspects of the present invention, and is characterized in that, when receiving a failure notification message in which a tunnel-directed label used by a service is a header, a corresponding service-directed label or a service identification is obtained from the header, and the user is notified of a disconnection state of the service.

The eighth aspect of the present invention provides a communication system according to any of the first to fifth aspects of the present invention; it is a failure notification system at an MPLS path endpoint and a connecting point in a transmission system. The transmission system connects paths in tandem formed by mapping to MPLS and provides bidirectional services on an end-to-end basis. The communication system is characterized in that when the MPLS path endpoint and the connecting point receive the failure notification message in which a tunnel-directed label used by a customer service is a header, a service-directed label in the forward direction is obtained from the header, and a disconnection state of the service is reflected in a payload associated with the obtained service-directed label.

The ninth aspect of the present invention provides a communication system according to any of the first to fifth aspects of the present invention, and is characterized in that, when the means to associate, on the basis of the received message used in the assignment control, associates the service-directed label included in the corresponding message with the received tunnel-directed label, if it is determined that all service-directed labels included in the corresponding message are present in a single tunnel which holds an opposite service, then the tunnel is notified that a service using a tunnel to which the received failure notification message belongs is in a disconnection state.

The tenth aspect of the present invention provides a communication system which is used in a transmission system to use MPLS (Multi Protocol Label Switching) tunneling and label technology to map a communication service to the MPLS and to provide bidirectional services on an end-to-end basis. The communication system is characterized by including a means to perform label assignment control for the bidirectional services; a means to receive, after completion of the assignment control, a message in which a message used in the assignment control is a payload and a tunnel-directed label used by the bidirectional services is a header; and a means to associate, on the basis of the received message used in the assignment control, a service-directed label included in the corresponding message with the received tunnel-directed label.

FIG. 1 is a model diagram of PW based on RFC3915 in the prior art;

FIG. 2A illustrates a prior art and an OAM structure adopted in Y.1731;

FIG. 2B illustrates a prior art and an OAM structure adopted in Y.1731;

FIG. 2C illustrates a prior art and an OAM structure adopted in Y.1731;

FIG. 3 illustrates a problem and is a schematic diagram of the operation in the conventional alarm processing in the case of using PW (Pseudo Wire) technology based on MPLS;

FIG. 4 is a schematic diagram illustrating the problem more specifically;

FIG. 5 is a diagram to illustrate a first example of the present invention;

FIG. 6A is a schematic diagram illustrating a processing of an embodiment of the present invention;

FIG. 6B is a schematic diagram illustrating a processing of an embodiment of the present invention;

FIG. 6C is a schematic diagram illustrating a processing of an embodiment of the present invention;

FIG. 7 is an apparatus block diagram of the first example of the present invention;

FIG. 8 is an apparatus block diagram of the first example of the present invention;

FIG. 9 is a diagram illustrating operations in the first example of the present invention with reference to FIGS. 6A , 6 B, 6 C, and 7 ;

FIG. 10A is a schematic diagram of the second example of the present invention;

FIG. 10B is a schematic diagram of the second example of the present invention;

FIG. 10C is a schematic diagram of the third example of the present invention;

FIG. 11 is an apparatus block diagram of the third example of the present invention; and

FIG. 12 is a schematic diagram of the fourth example of the present invention;