Error detection/correction and fault detection/recovery – Pulse or data error handling – Transmission facility testing
Reexamination Certificate
1996-03-13
2001-05-15
Pan, Daniel H. (Department: 2183)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Transmission facility testing
C709S200000, C709S203000, C709S208000, C709S210000, C709S251000, C714S004110, C714S043000, C714S056000, C714S712000
Reexamination Certificate
active
06233704
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to computer systems and in particular to highly interconnected dual ring computer network systems.
BACKGROUND
Modular, highly interconnected computer network systems offer the ability to exchange data among individual nodes in the network and to share hardware resources. Due to the large amount of data movement in such a system, communication traffic among the nodes can become congested. Some methods for reducing congestion have been proposed. One, the Scalable Coherent Interface increases throughput through the network by sending data in self-contained packets which have the needed address, command and data. That way, processors don't have to wait for each message to travel the length of the network before sending the next one. The Scalable Coherent Interface (SCI) is an IEEE Standard (IEEE Std 1596-1992). [David B. Gustavson, The Scalable Coherent Interface and Related Standards Projects, IEEE Micro, February 1992, p. 10].
The Scalable Coherent Interface is a single ring network which can be implemented on a range of systems, from smaller personal computer based networks to huge, interconnected network systems. Since it is a single ring system, however, the network is not tolerant to failures in the network hardware. For example, although the Scalable Coherent Interface has the feature that a blocked node only blocks packets destined to that node while not blocking packets destined elsewhere, this protocol relies on having a fully connected ring network. Breaking the ring in order to add, to remove, or to swap a node interrupts the Scalable Coherent Interface.
On the other hand, the Fiber Distributed Data Interface (FDDI) provides fault tolerance by using a dual-ring configuration. One of these rings is used for the data traffic. The secondary ring can be used to fold the communications ring around a faulty network node. This allows the network to remain functional without needing to physically replace the faulty node.
Even on a dual ring network such as FDDI, the process of testing faulty nodes in the network can be troublesome. As long as the faulty node remains on the network, that node can cause damage both to data kept on network storage devices and to the communication flowing through the node on the network.
There is a need in the art for a ring-based computer network which can increase the transfer rate of data. At the same time, there is a need for a computer network that can expand in a cost-efficient manner and one which is highly fault tolerant. This network should allow faulty nodes to be partially disabled so that the network is not harmed by such faults and so that the nodes can be tested and debugged without needing to remove them from the network. There should also be the ability to add, remove, or swap nodes while the rest of the network can continue to function normally.
SUMMARY OF THE INVENTION
The present invention is a computer network system with multiple, counter-rotating rings. The network is made up of nodes which consist of a client computer attached to the network by an interface device. The interface device provides input and output ports to the client computer as well as input and output ports to the network rings. The interface device has a plurality of buffers. One set, the receive buffers, accept data packets from the network to be passed along to the client computer. The other set, the send buffers, accumulate and send data packets from the client computer out to the network. The interface device also has a plurality of bypass buffers and reversers. The bypass buffers allow the node to accept a data packet addressed to another node and hold the packet until it can be routed to the next neighboring node. The reversers allow the plurality of rings to be folded into larger rings in order to route network data around faulty nodes. Memory mapped registers are used to mask out entire rings when necessary.
The interface device also provides a client isolation device which allows the client computers to have multiple types of connectivity to the computer network. Client computers which are faulty, for example, can be isolated and given very limited connectivity. On the other hand, a client computer acting as a network supervisor, can be given full connectivity to the network.
REFERENCES:
patent: 4803620 (1989-02-01), Inagami et al.
patent: 4845710 (1989-07-01), Nakamura et al.
patent: 4866704 (1989-09-01), Bergman
patent: 5191656 (1993-03-01), Forde, III et al.
patent: 5195100 (1993-03-01), Katz et al.
patent: 5233618 (1993-08-01), Glider et al.
patent: 5274799 (1993-12-01), Brant et al.
patent: 5289470 (1994-02-01), Chang et al.
patent: 5303244 (1994-04-01), Watson
patent: 5345565 (1994-09-01), Jibbe et al.
patent: 5381534 (1995-01-01), Shi
patent: 5390188 (1995-02-01), Dawson
patent: 5402428 (1995-03-01), Kakuta et al.
patent: 5432784 (1995-07-01), Ozveren
patent: 5452330 (1995-09-01), Goldstein
patent: 5452444 (1995-09-01), Solomon et al.
patent: 5457683 (1995-10-01), Robins
patent: 5499275 (1996-03-01), Kishi
patent: 5499337 (1996-03-01), Gordon
patent: 5513192 (1996-04-01), Janku et al.
patent: 5519844 (1996-05-01), Stallmo
patent: 5522031 (1996-05-01), Ellis et al.
patent: 5530948 (1996-06-01), Islan
patent: 5590124 (1996-12-01), Robins
patent: 5594869 (1997-01-01), Hawe et al.
patent: 5617409 (1997-04-01), Ozveren et al.
patent: 5617570 (1997-04-01), Russell et al.
patent: 5619645 (1997-04-01), Montenegro et al.
patent: 5638448 (1997-06-01), Nguyen
patent: 5659721 (1997-08-01), Shen et al.
patent: 5751696 (1998-05-01), Bechtel et al.
patent: 5838993 (1998-11-01), Riley et al.
Gustavson, D B, “the Scalable Coherent Interface and Related Standards Projects”,IEEE Micro,10-22, (Feb., 1992).
“IEEE Standard for Scalable Coherent Interface (SCI)”,IEEE Std 1596-1992,1-248, (Mar., 1992).
Patterson, D A, et al., “A Case For Redundant Arrays Of Inexpensive Disks (RAID)”,University of California at Berkeley, Report No. UCB/CSD 87/391,(Dec. 1987).
Scott,S, “The SCX Channel: A New, Supercomputer-Class System Interconnect”,HOT Interconnects III,Abstract, pp. 1-11, (Aug. 1-11, 1995).
Kunkel Daniel L.
Oberlin Steven M.
Schwoerer Gerald A.
Scott Steven L.
Nguyen Dzung C
Pan Daniel H.
Schwegman Lundberg Woessner & Kluth P.A.
Silicon Graphics Inc.
LandOfFree
System and method for fault-tolerant transmission of data... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for fault-tolerant transmission of data..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for fault-tolerant transmission of data... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2542058