Bi-directional chaining of network access ports

Details Bi-directional chaining of network access ports Bi-directional chaining of network access ports

1999-12-27

2004-01-20

Ho, Duc (Department: 2665)

Multiplex communications

Fault recovery

Bypass an inoperative switch or inoperative element of a...

C370S352000, C370S392000, C370S395100, C370S465000

Reexamination Certificate

active

06680904

ABSTRACT:

FIELD OF THE INVENTION
The present invention relates generally to high-speed data communication systems, and specifically to digital subscriber line systems.
BACKGROUND OF THE INVENTION
Digital Subscriber Line (DSL) is a modem technology that enables broadband digital data to be transmitted over twisted-pair wire, which is the type of infrastructure that links most home and small business subscribers to their telephone service providers. DSL modems enable users to access digital networks at speeds tens to hundreds of times faster than current analog modems and basic ISDN service. DSL thus opens the most critical bottleneck in local-loop access to high-speed networks, such as Asynchronous Transfer Mode (ATM) and Internet Protocol (IP) networks, without requiring major investments in new infrastructure. A range of DSL standards have been defined, known generically as “xDSL,” wherein the various standards have different data rates and other associated features but share common principles of operation.
DSL subscribers are connected to high-speed networks through Digital Subscriber Line Access Multiplexer (DSLAM) systems. Because of the high cost of network bandwidth, a single DSLAM must typically be designed to serve between 100 and 1000 subscribers and to concentrate their traffic through one or a few network trunks. The need to serve such a large and potentially variable number of subscribers in the one DSLAM has led to the development of “multi-shelf” access architectures.
FIG. 1
is a block diagram that schematically illustrates a DSLAM system
20
implementing a multi-shelf architecture, as is known in the art. System
20
comprises a master unit
25
coupled to multiple slave units
27
in a star configuration. Master unit
25
communicates with a core network
22
, such as an ATM network. The master unit comprises a core network interface element
24
, providing the necessary physical layer (PHY) and data link layer (for example, ATM) functions. A concentrator
26
performs higher-level functions, including VPI/VCI translation for the ATM network, and multiplexes downstream and upstream packets, or cells, among slave units
27
. Each slave unit typically comprises a switching core
29
, coupled to a plurality of ports
28
serving respective subscriber premises via suitable DSL modems. In the physical implementation of system
20
, each such slave unit occupies its own shelf in an electronics rack, and this is the reason for the use of the term “multi-shelf.”
FIGS. 2A and 2B
are block diagrams that schematically illustrate topologies known in the art for use in multi-shelf access systems, such as that shown in FIG.
1
.
FIG. 2A
shows a simple star configuration. In a redundant star configuration, shown in
FIG. 2B
, each slave unit is connected to two or more alternative master units, so as to provide protection in case of a failure in one of the master units.
SUMMARY OF THE INVENTION
It is an object of some aspects of the present invention to provide an improved multi-shelf access architecture.
It is a further object of some aspects of the present invention to provide apparatus and methods for efficient, high-speed transfer of data packets within an access multiplexer system.
It is still a further object of some aspects of the present invention to improve the robustness and reliability of multi-shelf access systems.
In preferred embodiments of the present invention, a multi-shelf access system comprises two master units (“masters”) and a plurality of slave units (“slaves”) One of the masters serves as an active master, through which the slaves normally receive data packets from a core network and transmit packets back upstream to the network. The other master is termed a standby master. The slaves are connected in one or more daisy chains between the active and standby masters, and are configured so that both downstream and upstream packets can be transmitted in either direction along each of the chains, i.e., from or to either of the two masters. Thus, if a failure occurs in any one of the slaves or in a link between them, the traffic direction in the chain in which the failure has occurred is simply reversed so as to run through the standby master. Substantially no reprogramming of the slaves is required for this purpose. All of the slaves (except possibly a slave that has failed) continue to provide service to subscribers, typically with only a short interruption until corrective action is completed.
Preferably, the upstream packets are bicast by the active master to both the core network and to the standby master. The standby master also transmits the upstream packets to the core network, in order to protect against faults in the network interface of the active master unit.
In some preferred embodiments of the present invention, each slave comprises a pre-switch, which examines each packet passed down its respective chain to determine whether the packet is destined for any of the ports on that slave. Preferably, every packet passed between the masters and slaves has a pre-switch address. If the packet is addressed to a particular slave, the pre-switch on that slave conveys the packet to a switch or switching fabric of the slave, which routes the packet to the appropriate port. If the packet is a multicast packet with an appropriate pre-switch address, the pre-switch both conveys a copy of the packet to the switch or switching fabric of the slave and passes the packet on down the chain. The pre-switch similarly accepts packets carrying a “next” address, i.e., packets that are sent from one slave to its immediate neighbor for topology detection and diagnostics. Otherwise, the pre-switch rejects the packet and passes it on to the next slave along the chain.
The pre-switch thus reduces the burden on the switching fabric of the slave and ensures that packets are passed down the chain at the full data rate supported by the master-slave interface. Furthermore, in the case of a fault in the slave, the pre-switch simply passes all packets through to the next slave in the chain, so that the effect of the fault is limited as far as possible.
Preferably, the slave pre-switch is bidirectional, i.e., it treats packets passed both up and down the chain in substantially the same manner. The pre-switch thus supports the failure protection mechanism described hereinabove, wherein the traffic direction is reversed in response to a failure in the chain.
There is therefore provided, in accordance with a preferred embodiment of the present invention, network access apparatus, including:
first and second master units, each including a physical interface to a packet-switched network;
a plurality of slave units, each slave unit including one or more ports to respective subscriber lines; and
a plurality of physical interface lines, which link the slave units in one or more daisy chains, in which the slave units are mutually connected in series by the physical interface lines therebetween, each daisy chain including at least a first slave unit connected by one of the physical interface lines to the first master unit and a last slave unit connected by another of the physical interface lines to the second master unit.
Preferably, in normal operation, downstream data packets received from the network are passed from the first master unit to each of the daisy chains via the first slave unit in each chain, and upstream data packets received by the slaves in each chain from the subscriber lines are passed via the first slave unit in the chain to the first master unit for transmission over the network. Further preferably, the apparatus includes a protection interface, which couples the second master unit to the first master unit, and over which interface data packets are conveyed between the first and second master units in case of a fault. Most preferably, the first master unit bicasts the upstream data packets that it receives from the slave units to the network and, via the protection interface, to the second master unit, which transmits the upstream data packets to the network.
Preferably, in c

Affiliated with

Also associated with

Rating

Bi-directional chaining of network access ports does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Bi-directional chaining of network access ports, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bi-directional chaining of network access ports will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3222007