Reliable local data network arrangement

• US 4,692,918 A
• Filed: 12/17/1984
• Issued: 09/08/1987
• Est. Priority Date: 12/17/1984
• Status: Expired due to Term

• Current Assignee: Bell Telephone Laboratories, Inc. (Nokia Corporation)
• Original Assignee: AT&T, Inc.
• Inventors: Elliott, Roger A., Lindquist, Wayne V.
• Primary Examiner(s): Olms, Douglas W.
• Assistant Examiner(s): Rokoff, Kenneth I.
• Application Number:
  US06/682,096
• Time in Patent Office: 995 Days
• Field of Search: 370/85, 370/89, 370/94, 370/16, 340/825.5
• US Class Current: 370/401
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• CPC Class Codes: G06F 11/22 Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
• CPC Class Codes: H04L 12/40 Bus networks
• CPC Class Codes: H04L 12/40032 Details regarding a bus interface enhancer
• CPC Class Codes: H04L 12/413 with random access, e.g. carrier-sense multiple-access with collision detection [CSMA-CD]
• CPC Class Codes: H04L 12/5692 Selection among different networks
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: H04M 3/241 for stored program controlled exchanges

Abstract

The reliability of a distributed processing system is enhanced through the use of a highly reliable data communication system. Each node of such a system is connected to the other nodes by a local data network, including duplicated data network media, such as coaxial cables. Each node includes a maintenance processor for controlling which of the duplicate network media is active and which is standby. Regular inter-node data packets are transmitted over the active network medium. Control data packets to the maintenance processor are sent from a control node over the standby network medium. The maintenance processor can also totally stop transmission of data from the associated node to both data network media.

Claims

1. In a distributed processing system comprising a plurality of processor complexes, each complex comprising at least one processor, means for transmitting data packets among said processor complexed comprising:

two local data network transmission media; and

a network interface uniquely associated with and connected to each of said processor complexes and connected to both of said transmission media for conveying operational data packets between the associated processor complex and one of said transmission media and comprising a maintenance processor for receiving control data packets from the other of said transmission media and responsive to received control data packets for controlling maintenance operations of said network interface.

2. In a distributed processing system comprising a plurality of processor complexes, each complex comprising at least one processor, means for transmitting data packets among said processor complexes comprising:

two local data network transmission media; and

a network interface uniquely associated with and connected to each of said processor complexes and connected to both of said transmission media for conveying operational data packets between the associated processor complex and one of said transmission media and for receiving control data packets from the other of said transmission media, wherein said network interface comprises a maintenance processor responsive to said control data packets for selecting said one and said other of said transmission media.

3. A network interface for use in a distributed processing system, said distributed processing system comprising two local data network transmission media and a plurality of processor complexes, each complex comprising at least one processor, said network interface associated with and connected to one processor complex and connected to both of said transmission media, one of said transmission media being designated active and one of said transmission media being designated standby, for transmitting data packets between said one processor complex and said media, comprising:

a maintenance processor for receiving control packets from the one of said transmission media designated standby and, responsive to receipt of said control packets, for changing the designation of one of said transmission media to active and changing the designation of one of said transmission media to standby; and

means for conveying data packets between said one processor complex and the one of said transmission media designated active by said maintanence processor.

4. The network interface of claim 3 in which said maintenance processor comprises means for generating output control signals to control said means for conveying data packets to transmit to no more than one of said two transmission media.

5. The network interface of claim 4 in which said means for generating said output control signals is further adapted to control said means for conveying data packets to transmit to neither of said two transmission media.

6. The network interface of claim 4 in which said maintenance processor is further responsive to said control packets, received from the one of said transmission media designated standby, to generate said output control signals.

7. In a distributed processing system, comprising a plurality of processor complexes and data communication means interconnecting said processor complexes, a method of communicating data between a first and a second of said processor complexes, comprising the steps of:

providing two local data network transmission media as part of said data communication means;

designating one of said transmission media as active and the other of said transmission media as standby

transmitting data packets from said first processor complex over said active transmission medium;

receiving said data packets in said second processor complex from said active transmission medium;

transmitting control packets from said second processor complex over said standby transmission medium; and

receiving said control packets in said first processor complex from said standby transmission medium.

8. The method of communicating data of claim 7 in which said step of receiving said control packets further comprises the steps of:

generating an interrupt signal to said first processor complex.

9. The method of communicating data of claim 7, further comprising the step of:

responsive to reception of said control packets in said first processor complex, changing the designation of said one transmission medium to standby and the designation of said other transmission medium to active.

10. In a distributed processing system comprising a plurality of processor complexes, each complex comprising at least one processor, means for transmitting data packets among said processor complexes comprising:

two local data network transmission media; and

a network interface uniquely associated with and connected to each of said processor complexes and connected to both of said transmission media, comprising a maintenance processor for designating one of said transmission media active, for designating one of said transmission media standby, and means for conveying data packets between the associated processor complex and the one of said transmission media designated active by said maintenance processor.

11. The distributed processing system of claim 10 in which said maintenance processor comprises means for generating output control signals to control said means for conveying data packets to transmit to no more than one of said two transmission media.

12. The distributed processing system of claim 11 in which said means for generating said output control signals is further adapted to control said means for conveying data packets to transmit to neither of said two transmission media.

13. The distributed processing system of claim 12 in which said maintenance processor is further adapted to designate one of said transmission media as standby, and is responsive to control packets, received from the one of said transmission media designated standby, to generate said output control signals.

14. The distributed processing system of claim 11 in which said maintenance processor is responsive to control packets, received from the one of said transmission media designated standby, to generate said output control signals.

15. The distributed processing system of claim 10 in which said maintenance processor is further adapted to receive control packets from the one of said transmission media designated as standby by said maintenance processor.

16. The distributed processing system of claim 15 in which one of said plurality of processor complexes and its associated network interface is designated a control node, and in which said control node is operative under program control to generate and transmit said control packets on the one of said transmission media designated standby by the maintenance processor of said control node.

17. The distributed processing system of claim 16, in which the network interface of said control node is responsive to data received from a processor of the processor complex of said control node to convey said control packets from said processor complex of said control node to the one of said transmission media designated standby by the maintenance processor of said control node.

18. The distributed processing system of claim 15 in which said network interface further comprises means for receiving said control packets from the one of said transmission media designated active by said maintenance processor and transmitting said control packets to said maintenance processor.

19. The distributed processing system of claim 10 in which one of said plurality of processor complexes and its associated network interface is designated a control node, and in which said control node is operative under program control to transmit control packets on the one of said transmission media designated standby by the maintenance processor of said control node.

20. The distributed processing system of claim 19, in which the network interface of said control node is responsive to data received from a processor of the processor complex of said control node to convey said control packets from said processor complex of said control node to the one of said transmission media designated standby by the maintenance processor of said control node.

21. The distributed processing system of claim 10 in which said network interface further comprises an interface processor for controlling said means for conveying data.

22. The distributed processing system of claim 21 in which said maintenance processor is responsive to a control packet received over said one of said transmission media designated as standby to generate an interrupt signal to said interface processor.

23. The distributed processing system of claim 22 in which said maintenance processor is further adapted to generate data signals, and in which said interface processor is responsive to said interrupt signal and said data signals to initiate a program identified by said data signals.

24. The distributed processing system of claim 10 in which said maintenance processor is responsive to a control packet received over said one of said transmission media designated as standby to generate an interrupt signal to one of the processors of the processor complex associated with said network interface.

25. The distributed processing system of claim 24 in which said maintenance processor is further adapted to generate data signals, and in which said one of the processors of the processor complex associated with said network interface is responsive to said interrupt signal and said data signals to initiate a program identified by said data signals.

26. The distributed processing system of claim 10 wherein said maintenance processor is further responsive to data signals received from said associated processor complex to initiate execution of a program.

27. The distributed processing system of claim 10 wherein said network interface comprises an interface processor for controlling said means for conveying data.

28. The distributed processing system of claim 27 wherein said maintenance processor is further responsive to data signals received from said interface processor to initiate execution of a program.