Multiplex communications – Fault recovery
Reexamination Certificate
1997-08-15
2001-05-01
Olms, Douglas W. (Department: 2732)
Multiplex communications
Fault recovery
C370S396000, C370S401000, C370S410000
Reexamination Certificate
active
06226260
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to computer internetworking. More particularly, the present invention is directed to a resilient internetwork connection between frame relay networks.
2. Related Art
Frame relay (FR) end user-systems or networks are interconnected by a separate data network. User-network interfaces link each FR end user-system to the data network. In commercial applications, these user-network interfaces include gateway switches provided in nodes near business premises for switching data between a FR network and the interconnecting data network. Greater volumes of vital communications traffic are now sought to be sent over long-distances between interconnected FR systems. Sound network-to-network interconnections which transfer data reliably and quickly from end-to-end without error is increasingly important.
Unfortunately, a number of problems can occur resulting in network-to-network failure. Physical trunk failures, such as a fiber cut, can halt communication between frame relay networks. For example, lightning, backhoes, train derailment, trawler nets, and shark attacks have been reported to have severed or impaired fiber connectivity. See, Grover, W., PhD., “Distributed Restoration of the Transport Network,” p 337. Even more likely culprits for internetwork connection error are hardware and/or software created problems such as network-interface card (NIC) failure or human error in software loading.
Self-healing networks, such as asynchronous transfer mode (ATM) networks, can restore communication to overcome some types of failure. Restoration algorithms and protocols allow the self-healing network to re-route data within the self-healing network to avoid a faulty switch, node, or trunk. Even failures at a physical user-network interface connected to the self-healing network can be detected. A communication path within the self-healing network can be restored by switching to a pre-configured back-up path through the self-healing network.
Unfortunately, such resiliency of a self-healing network is limited to an internal response within the self-healing network. Without communication from the self-healing network, end-user systems remain unaware of a detected failure. Failures at the user-network interface are especially troublesome. For example, a broken physical interface linking an end user system and self-healing network will be detected and responded to by the resilient network. The ignorant end-user system, however, will continue to try to send and receive data through the failed physical interface. In this case, the Network-Network Interface (NNI) between the end-user network and the self-healing network is not resilient.
This problem is especially acute between interconnected frame relay networks. Frame Relay is only specified as a User-to-Network Interface (UNI) protocol. There is no standard protocol specified between a FR network and a self-healing network which allows a Network-Network Interface (NNI) to automatically detect a single point failure such as failed physical interface (or link). End-to-end performance suffers as the NNI cannot re-route traffic over non-failed or pre-configured back-up link(s). Thus, the lack of NNI resiliency prevents the interconnection of frame relay networks with end-to-end reliability which is not vulnerable to a single point failure.
What is needed then is a resilient interconnection from end-to-end between frame relay networks. A resilient NNI is needed which can re-route traffic to avoid a failed physical interface or link. For FR networks interconnected by a self-healing network, a standard protocol is needed which allows a NNI to automatically respond to physical interface failures detected by the self-healing network.
SUMMARY OF THE INVENTION
The present invention pertains to a method and system for a resilient internetwork connection of frame relay (FR) networks. For FR networks interconnected by a self-healing network, a standards-based protocol is provided which allows a resilient network-to-network interface (NNI) to automatically respond to physical interface failures detected by the self-healing network. In the resilient NNI, switching gateways re-route further traffic to avoid the failed physical interface. End-to-end resiliency is obtained between frame relay networks interconnected by a self-healing network which is not vulnerable to single point of failure.
In one embodiment of the present invention, a resilient NNI is provided between a FR network and a self-healing ATM network. A simple set of interacting protocols and mechanisms is provided between FR/ATM gateways and ATM switches. Each FR/ATM gateway includes interworking function (IWF) processing modules for converting user data between frame relay packets and ATM cells and for processing status signaling messages to obtain a resilient NNI. The status signaling messages are used to determine when the physical interface which receives cell data differs from the physical interface over which cell data was sent. This change in the receive physical interface indicates the self-healing network has switched from using the transmit physical interface due to a detected failure within the self-healing network, or at the destination gateway.
In one FR/ATM embodiment, each FR/ATM gateway includes a switch interconnecting FR-IWF and ATM-IWVF modules. According to the present invention, Tx ATM PHY and Rx ATM PHY fields are used internally within the FR/ATM gateway to identify the ATM physical interfaces through which cells are transmitted and received from the ATM network. Through status signaling messages, each ATM-IWF module transmits a Rx ATM PHY field in response to receiving ATM cells from the ATM network. Each FR-IWF module then compares Tx ATM-PHY and Rx ATM-PHY fields to determine whether the receive ATM physical interface differs from the transmit ATM physical interface as a result of self-healing in the ATM network. Cell traffic can then be switched to the receive ATM physical interface or to another pre-configured back-up physical interface.
Thus, the present invention takes advantage of the automatic self-healing in the ATM network by including interworking processing at each FR/ATM gateway to detect differences between transmit physical interfaces and receiver physical interfaces for each logical connection trunk. In this way, each FR/ATM gateway in response to status signaling messages can switch rapidly and accurately from a primary ATM UNI physical link to a back-up link. A resilient end-to-end frame relay system is realized which overcomes single-point failures, and even some dual failures in the interconnection network.
In general, a resilient NNI according to the present invention can be provided in any two FR switching nodes linked by a self-healing network, which may be as simple as two pairs of physical connections between the gateways. A four node full mesh design can also be realized. A further simple extension allows a two-level hierarchical network of resilient NM nodes to be implemented that can scale to very large sizes via a tandem FR/ATM switching function. Further hierarchical switching can employ the Virtual Path switching concept of ATM to support even larger networks.
Further features and advantages of the present invention, as well as the structure and operation of various embodiments of the present invention, are described in detail below with reference to the accompanying drawings.
REFERENCES:
patent: 5105420 (1992-04-01), Ardon et al.
patent: 5473679 (1995-12-01), LaPorta et al.
patent: 5490140 (1996-02-01), Abensour et al.
patent: 5502712 (1996-03-01), Akita
patent: 5600798 (1997-02-01), Cherukuri et al.
patent: 6094419 (2000-07-01), Ohyoshi et al.
patent: 0 484 943 A2 (1992-05-01), None
Cohen et al., “Connection Management and Rerouting in ATM Networks,”Proceedings of the Conference on Computer Communications, vol. 1, Jun. 12, 1994, pp. 184-191.
Shinbo et al., “Frame Relay and ATM Network Interworking for High Speed Public Data Communication Networks,”Pr
MCI Communications Corporation
Olms Douglas W.
Vanderpuye Ken
LandOfFree
Method and system for resilient frame relay network... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for resilient frame relay network..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for resilient frame relay network... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2560648