Multi-layer network device in one telecommunications rack

Details Multi-layer network device in one telecommunications rack Multi-layer network device in one telecommunications rack

2000-06-16

2004-07-06

Ho, Duc (Department: 2665)

Multiplex communications

Pathfinding or routing

Switching a message which includes an address header

C370S469000, C361S788000, C709S249000, C710S120000

Reexamination Certificate

active

06760339

ABSTRACT:

BACKGROUND
Within typical telecommunications (“telco”) networks, network packets are transferred between network devices over a physical attachment or connection in accordance with a particular physical layer (i.e., layer 1) protocol, for example, Ethernet or Synchronous Optical NETwork (SONET). The physical layer may include optical fibers, coaxial cable and twisted wire. The physical layer is expensive because of the installation and material costs required to span the long distances that must be covered as well as the costs associated with the physical/port interfaces required to connect to the cable, wire or fiber.
The network packets transferred across the physical layer include both data and control information, and the data within each network packet may be further organized in accordance with an upper layer network protocol including, for example, Asynchronous Transfer Mode (ATM), Internet Protocol (IP), Multi-Protocol Label Switching (MPLS), Frame Relay, Voice, Circuit Emulation or Optical User Network Interfaces (UNI), including Optical Internetworking Forum (OIF) and Optical Domain Service Interconnect (ODSI). Although each physical attachment may be used to transfer network packets having data organized in accordance with multiple upper layer protocols, layer 1 network devices are typically not aware of which upper layer protocols are being carried. Instead, connections are set up manually through layer 1 network devices, such as electronic Digital Access and Cross-connect Systems (DACS) or optical cross-connect systems, and the connections cause the cross-connect systems to separate the network packets in accordance with each upper layer protocol—often referred to as “grooming”—and send them to appropriate upper layer network devices over external network attachments.
Once separated, the network packets (or just the data from within the network packets) are transferred to separate physical interfaces—corresponding to the different upper layer protocols—on one or more separate upper layer network devices. For example, network packets may be sent in accordance with the SONET protocol but include data organized in accordance with ATM (i.e., ATM over SONET) and MPLS (i.e., MPLS over SONET). Through established connections the cross-connect system transfers network packets including data organized in accordance with ATM to one or more physical interfaces supporting ATM and network packets including data organized in accordance with MPLS to one or more other physical interfaces supporting MPLS. The physical interfaces may be on one or more network devices, such as switches, routers and hybrid switch-routers, where hardware subsystems within the network device(s) and connected to each physical interface are capable of operating on network packet data in accordance with a particular upper layer protocol. The transport or layer 1 connections between the cross-connect system and the upper layer devices generally take considerable time to set up and are not easily or dynamically changed.
The cross-connect system and the upper layer devices are often owned by different parties, in which case, the owner of the cross-connect system generally charges a “grooming” fee to the owner(s) of the upper layer devices to which it is connected. In addition, due to the different functionality of a physical layer network device, such as a cross-connect system, and an upper layer network device, such as an ATM switch, and the complexity of each different device, different engineering skill sets are required to develop each type of device and generally different vendors develop the different network devices.
Within a typical telecommunications network, network devices for connecting different portions of the network are located at various telecommunications sites—often referred to as central offices. There may be hundreds of devices per site and, thus, space is a valuable resource. Consequently, network devices must conform to the size of a standard telecommunications rack (“telco rack”)—that is, the device must fit within a rack no taller than seven feet, no wider than nineteen inches and no deeper (often referred to as the “line up”) than twenty-four inches. These dimensions may vary slightly.
Under deregulation, telco site owners must lease space to certain other carriers. These leases are expensive and increase with the amount of space leased. In addition, there are limits to the amount of space that must be leased to each carrier.
In general, the components of the network device are located on multiple functional printed circuit boards, which must be coupled together in order for them to communicate and pass network packets. A back-plane or mid-plane provides such a coupling and is basically a printed circuit board having connectors into which each of the functional printed circuit boards may be plugged and electrical connections or etches between the different connectors through which the functional printed circuit boards may communicate and pass network packets. A back-plane is located toward the back of a network device or other computer and includes connectors for connecting to functional printed circuit boards on only one side—generally, the side facing the front of the device. A mid-plane is located toward the middle of a network device or other computer and includes connectors for connecting to functional printed circuit boards on both sides. Hereinafter, back-plane and mid-plane will be referred to as mid-plane for convenience.
The size of and number of routing layers within the printed circuit board of the mid-plane also sets limitations on network devices. The size of the back-plane or mid-plane determines the maximum number of connectors that may be located on the printed circuit board and the number of routing layers determines the maximum number of signals that may be routed between the connectors. Currently, the largest common size that a raw printed circuit board or panel comes in is three feet in length and, typically, only about thirty-four inches is usable. Larger sizes may be obtained but are exorbitantly expensive. Hence, the largest mid-plane that is likely to be used within a network device is about thirty-four by twenty-four inches. In addition, forty-eight routing layers is considered a very large number of routing layers and is likely to be the largest used within a network device. More routing layers, such as 52, are available but again, the price becomes exorbitant.
The thirty-four inch length of the mid-plane printed circuit board prevents a single network device from fully utilizing the entire seven feet available in a telco rack. In addition, the size and number of routing layers of the mid-plane printed circuit board also limits the number of ports that may be serviced by a single network device. As a result, typically two or more network devices are stacked within a teleco rack.
The need for network bandwidth has risen exponentially in recent years, and many new network devices have been developed in response to this need. Decreasing component sizes have allowed for increased port density (i.e., more ports per network device), and faster processors and other components have allowed for increased switching capacity (i.e., number of bits transferred per second). In 1995, the maximum switching capacity available in a single network device was five gigabits per second (Gbps). Switching capacity has doubled approximately every two years such that currently, the maximum switching capacity available in a single network device is 80 Gbps with 160 Gbps expected later this year and 320 Gbps expected by 2001. Despite decreasing component sizes, the physical constraints of both the telco rack and printed circuit boards limit the port density and switching capacity that may be provided by a single network device.
In an effort to provide network devices supporting increased bandwidth, several network device developers have developed network devices having portions or sub-sections located in two or more telco racks. These portions or sub-sections are interconnected thro

Affiliated with

Also associated with

Rating

Multi-layer network device in one telecommunications rack does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Multi-layer network device in one telecommunications rack, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Multi-layer network device in one telecommunications rack will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3224005