Electrical computers and digital processing systems: multicomput – Computer-to-computer data routing
Reexamination Certificate
1997-03-07
2003-07-08
Powell, Mark R. (Department: 2142)
Electrical computers and digital processing systems: multicomput
Computer-to-computer data routing
C370S401000
Reexamination Certificate
active
06591303
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of computer networks. More specifically, the present invention relates to a method and apparatus for parallel trunking of interfaces to increase transfer bandwidth between network devices.
BACKGROUND OF THE INVENTION
Local Area Networks (LANs) following the IEEE 802 Standard Architecture are network environments that interconnect end-nodes using various types of network elements. Each of these network elements is subject to its own set of configuration and topology guidelines. These sets of guidelines and rules are intended to deliver a uniform and well defined data link layer behavior over which upper network layers can operate. Examples of this behavior include the emulation of loop-free broadcast domains where any node may communicate with any other node with no prior signaling required, where packet ordering is preserved between any pair of end-nodes, and where every packet is delivered to its destination no more than once.
FIG. 1
illustrates a plurality of end-nodes interconnected by a network. End-nodes
110
-
113
are typically computers capable of sourcing packets to be delivered over the network
120
to other end-nodes. The end-nodes
110
-
113
are also capable of sinking packets sent to them by other end-nodes in the network
120
. Routers are also considered end-nodes for the purposes of this description. Typical network elements used to build LANs are repeaters (hubs) and bridges. In some cases bridges are designated as LAN switches. The IEEE 802.1d Standard for transparent bridges (LAN switches) and IEEE 802.3 Standard for repeaters and end-nodes provide the topological rules that allow the data link layer behavior described.
Any one of the end-nodes
110
-
113
may send packets to any other end-node connected to the network
120
. Packets are delivered at most once per destination. A sequence of packets sourced at one of the end-nodes
110
-
113
is seen in the same order at a destination end-node connected to the network
120
.
FIG. 2
illustrates a network implementation. The network
200
includes a plurality of repeaters
210
and
211
and switches
220
-
222
. As a rule, links or interfaces connected to repeaters have the same nominal speed or bit rate and links or interfaces connected to switches can have dissimilar speeds or bit rates. An additional property of the switches
220
-
222
is traffic isolation. Traffic isolation consists of forwarding packets of data only through the links or interfaces where the destination may reside. This property was deliberately defined to increase the overall network capacity. In order to accomplish traffic isolation, switches
220
-
222
must know which end-nodes
230
-
236
are reachable via each link or interface. Bridges or switches complying with IEEE 802.1d automatically learn this information with no assistance from the end-nodes
230
-
236
or any other external agent, and are called transparent bridges.
FIG. 3
illustrates a transparent bridge. Bridge
310
is connected to end-nodes
311
-
313
. A packet from end-node
311
to end-node
312
is not sent to the link where end-node
313
resides because of the property of traffic isolation. The bridge
310
learns the location of the end-nodes
311
-
313
.
Network capacity is central to the function of a network. Traditionally, there have been two approaches to increasing network capacity. The first approach involves partitioning the network. The partitioning approach uses switches, bridges, and routers to replace a network of repeaters. By replacing the network of repeaters with more sophisticated hub devices, the flow of packets in the network is better managed and system performance is increased. The second approach involves providing faster link alternatives. By implementing links that can support higher bandwidth, the network capacity is increased.
Nevertheless, at any given point in time, the choice of link speeds (10, 100, 1000 Mbps) may not match up very well with the amount of sustained throughput that a particular device can support. When switches and high performance routers are used to interconnect multiple links of a given speed, there is a clear need for the inter-switch or inter-router link to be able to support at least some aggregation of the links. If new hardware is required to utilize a newer, higher speed, increased bandwidth network, the utilization of the newer network may not be as attractive from a cost standpoint. Furthermore, the cost associated with implementing a newer network with more sophisticated hubs or faster links may also not be attractive from a cost standpoint.
Thus, what is needed is a method and apparatus for increasing data transfer bandwidth between network devices, such as end-nodes and switches.
SUMMARY
A method for interconnecting a first device and a second device in a network is described. The first device and the second device are connected to a plurality of interfaces. The plurality of interfaces emulate a single high-speed interface.
A method for creating a multiple interface connection is disclosed. A first identifier is assigned to a first interface and a second interface at the first device. A path between the first device to the second device is identified with the first identifier.
A second method for creating a multi-interface connection is disclosed. A first device is connected to a plurality of interfaces. The plurality of interfaces emulate a single high-speed interface.
A network is disclosed. The network includes a first device and a second device. A first interface is coupled to the first device and the second device. A second interface is coupled to the first device and the second device. The first interface and the second interface emulate a single high speed interface. According to an embodiment of the present invention, the first interface and the second interface are assigned an identifier that identifies a path between the first device and the second device.
A network device is disclosed. The network device includes a first port that is connected to a first interface. The network also includes a second port that is connected to a second interface. A trunking pseudo driver is coupled to the first port and the second port. The trunking pseudo driver allows the first interface and second interface to emulate a single high-speed device.
REFERENCES:
patent: 4703475 (1987-10-01), Dretzka et al.
patent: 5051926 (1991-09-01), Stevens et al.
patent: 5379296 (1995-01-01), Johnson et al.
patent: 5390232 (1995-02-01), Freeman
patent: 5420857 (1995-05-01), Jurkevich
patent: 5517488 (1996-05-01), Miyazaki et al.
patent: 5537394 (1996-07-01), Abe et al.
patent: 5546549 (1996-08-01), Barrett et al.
patent: 5550816 (1996-08-01), Hardwick et al.
patent: 5586261 (1996-12-01), Brooks et al.
patent: 5619183 (1997-04-01), Ziegra et al.
patent: 5636371 (1997-06-01), Yu
patent: 5689508 (1997-11-01), Lyles
patent: 5719870 (1998-02-01), Baker et al.
patent: 5724356 (1998-03-01), Parameswaran Nair et al.
patent: 5732071 (1998-03-01), Saito et al.
patent: 5774668 (1998-06-01), Choquier et al.
patent: 5781854 (1998-07-01), Bales
patent: 5805586 (1998-09-01), Perreault et al.
patent: 5805816 (1998-09-01), Picazo, Jr. et al.
patent: 5815505 (1998-09-01), Mills
patent: 5896371 (1999-04-01), Kobayashi et al.
patent: 5923667 (1999-07-01), Poiraud et al.
patent: 5933412 (1999-08-01), Choudhury et al.
patent: 5951650 (1999-09-01), Bell et al.
patent: 6049528 (2000-04-01), Hendel et al.
patent: 6064671 (2000-05-01), Killian
patent: 6081833 (2000-06-01), Okamoto et al.
patent: 0685952 (1995-12-01), None
Browning, Optimal Stochastic Flow Control in a High Speed Virtual Circuit, IEEE 1993.*
Ko et al, Optimal End-to-ENd Sliding Window Flow Control in High-Speed Networks, IEEE 1991.*
Data Communications, vol. 24, No. 13, Sep. 21, 1995, pp. 85-88, 90.
IBM Technical Disclosure Bulletin, vol. 38, No. 9, Sep. 1, 1995, pp. 7-9.
Frazier Howard
Hejza Leo
Hendel Ariel
Blakely & Sokoloff, Taylor & Zafman
Powell Mark R.
Sun Microsystems Inc.
Vu Thong
LandOfFree
Method and apparatus for parallel trunking of interfaces to... 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 apparatus for parallel trunking of interfaces to..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for parallel trunking of interfaces to... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3076213