Multiplex communications – Pathfinding or routing – Switching a message which includes an address header
Reexamination Certificate
1998-07-24
2002-08-13
Ngo, Ricky (Department: 2664)
Multiplex communications
Pathfinding or routing
Switching a message which includes an address header
C370S409000, C370S466000
Reexamination Certificate
active
06434156
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates generally to computer networking. More specifically, the invention relates to switching, routing and bridging devices for interconnecting networks and other devices.
2. Description of the Related Art
One way of subdividing communications networking environments and protocols is to classify them as either connection-oriented or connectionless. In connectionless environment or protocol, the source and destination address is specified for each “packet” or unit of data such that the network resolves these addresses to determine a route for the packets to travel in. The IP or Internet Protocol is an example of a connectionless service. By contrast, in a connection-oriented environment, a source and destination set-up or establish a routing or connection that does not change from packet to packet. One example of a connection-oriented service is the PSTN (Public Switched Telephone Network) where one phone sets up a call to another phone and establishes a permanent circuit wherein voice data can travel back and forth.
Another way to classify networks is based upon their speed in transferring data and the geographical or physical range of the data transfer. Generally, in this classification, networks may be considered LANs (Local Area Networks) or WANs (Wide Area Networks). LANs are usually limited in geographical range (for instance, an office building) whereas WANs may span over hundreds of miles such between offices located in different cities. LANs and WANs may be either connection-oriented or connectionless. When WANs and LANs are connected to other LANs and WANs using different protocols special network devices known as gateways (or routers) are needed so that information can travel between them.
Typically these network devices are hardware mechanisms that can provide physical connectivity translation (e.g., between Ethernet and fiber) and/or protocol translation (e.g., between IPX and IP (Internet Protocol). With the proper design and programming of such network devices potentially any type of networks can be interconnected. When more than two networks or network devices are attached at some common interconnect, such as a frame relay cloud, devices known as switches regulate the passage of data from one network device to another.
Where a network device also needs to switch because different data have different destinations after passing through the device, switching and routing functions may need to be combined. Though switching and routing functions have often been combined by placing switching software/hardware within a network device, one problem from a customer standpoint is the disposition of legacy devices already present in the customer's network which may function well for their purpose and do not warrant an expensive replacement upgrade. Since such legacy devices may not be able to be programmed or retrofitted to provide a new function such as switching, to utilize legacy devices, switches are often added to a common connection point (such as a leased line) as separate hardware components. In a connection-oriented environment such as frame relay, this often implies that the “carrier” or provider of the physical line for the network must provide a separate additional line for the new network device. Though “virtual circuits” (VCs) may be provided by the carrier, these are only obtained at substantial cost and ordinarily not controlled or configurable by the customer. When the number of network devices grows, the cost becomes prohibitive since each additional carrier line is accompanied by monthly charges as well as maintenance.
Where physically separate switching hardware is used to interconnect two or more dissimilar network devices to another network, the hardware is often too expensive or cumbersome for the task at hand. For instance, most switches are built with a large number of ports, typically
16
, and thus are under-utilized where only a handful of network devices may need to be connected. Where the switching hardware is plugged in (for instance, by means of a chipset) to a network device, this integration is typically provided on the backplane of the router device and thus will conform to the protocol or standard of the backplane, for instance ATM (Asynchronous Transfer Mode) and be incapable of providing a translation between dissimilar protocols that may later need interconnection.
Where software switching is provided for the router, the software switch is typically not configurable once installed in the sense that it is typically not capable of peer-to-peer connectivity. For instance, since a router is designed to move data to a pre-specified network cloud to which it is connected, when two or more network devices are inter-connected for moving data to that pre-specified network, they are then unable to move data directly between each other. In order to do communicate with one another, they must move their data first “up” to that pre-specified network after which the pre-specified network moves the data back “down” to the other router. This problem is often stated as an inability of “users” of the same level (such as two routers) to directly communicate in a peer-to-peer fashion without passing first through an “upper user” (such as the pre-specified network). This scheme is very expensive in every facet since the latency, load upon the network and connection cost is increased while speed, reachability and even security may be decreased.
Thus, there is a need for virtual switching scheme that permits any network devices at any level in a network topology to communicate with each other directly without resorting to re-routing upstream and downstream when such routing is extraneous. Additionally, there is a need for a inexpensive switching solution that does not involve proprietary protocol recognition or hardware implementation.
SUMMARY OF THE INVENTION
What is disclosed is an apparatus comprising a first interface capable of sponsoring a first protocol, a first service translation layer converting data formatted in the first protocol into streams, the first service translation layer extracting and appending a first path identifier to the streams, a core responsive to the first service translator for obtaining the streams and determining a second path identifier therefrom, the core transferring the streams according to the second path identifier, a second service translation layer accepting the transferred streams, the second service translation layer converting the transferred streams into data formatted according to a second protocol, and a second interface for asserting the data formatted into the second protocol.
Also disclosed is a method comprising converting data formatted in first protocol into streams, uplinking the streams to a core module, the streams accompanied by a path identifier, transferring by the core of the streams to a destination in accordance with the path identifier, converting at the destination the streams into data formatted in a second protocol, and issuing the data formatted in the second protocol.
Also disclosed is a system comprising a virtual omni-user switching (VOSR) network device, a first network coupled to the VOSR, the VOSR accepting data from the first network, and a second network coupled to the VOSR, the VOSR converting the accepted data into a format compatible with the second network, the VOSR capable of switching data among the first and second networks and other networks coupled to the VOSR.
Also disclosed is a method comprising creating tunnel segments in a public network supplied data pipe, maintaining a tunnel mapping in private networks bridged by the data pipe, and switching data between devices in one of the private networks with devices in another of the private networks bridged by the data pipe, the tunnel mapping determining the switching, the switching performed over the data pipe in the tunnels in a fashion transparent to the public network.
REFERENCES:
patent: 5717691 (1998-02-01), Dighe et al.
patent: 6069890 (2000-05-0
Blakely , Sokoloff, Taylor & Zafman LLP
Ngo Ricky
Nortel Networks Limited
LandOfFree
Virtual switching for interconnected networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Virtual switching for interconnected networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Virtual switching for interconnected networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2887218