Central processing apparatus for fault-tolerant computing

• US 4,453,215 A
• Filed: 10/01/1981
• Issued: 06/05/1984
• Est. Priority Date: 10/01/1981
• Status: Expired due to Term

• Current Assignee: Stratus Computer, Inc. (Stratus Technologies Bermuda Holdings Ltd.)
• Original Assignee: Stratus Computer, Inc. (Stratus Technologies Bermuda Holdings Ltd.)
• Inventors: Reid, Robert
• Primary Examiner(s): Heckler, Thomas M.
• Application Number:
  US06/307,525
• Time in Patent Office: 978 Days
• Field of Search: 364/200 MS File, 364/900 MS File, 371/67, 371/68
• US Class Current: 714/11
• CPC Class Codes: G06F 11/10 Adding special bits or symbols to the coded information, e.g. parity check, casting out 9's or 11's
• CPC Class Codes: G06F 11/1625 in communications, e.g. transmission, interfaces
• CPC Class Codes: G06F 11/1641 where the comparison is not performed by the redundant processing components
• CPC Class Codes: G06F 11/20 using active fault-masking, e.g. by switching out faulty elements or by switching in spare elements
• CPC Class Codes: G06F 11/2007 using redundant communication media
• CPC Class Codes: G06F 13/374 using a self-select method with individual priority code comparator

Abstract

A fault-tolerant computer system provides information transfers between the units of a computing module, including a processor unit and a memory unit and one or more peripheral control units, on a bus structure common to all the units. Information-handling parts of the system, both in the bus structure and in each unit, can have a duplicate partner. Error detectors check the operation of the bus structure and of each system unit to provide information transfers only on fault-free bus conductors and between fault-free units. The computer system can operate in this manner essentially without interruption in the event of faults by using only fault-free conductors and functional units.

Arbitration circuits of unusual speed and simplicity provide units of the computing module with access to the common bus structure according to the priority of each unit.

The units of a module check incoming and outgoing signals for errors, signal other module units of a detected error, and disable the unit from sending potentially erroneous information onto the bus structure.

Claims

1. In central processing apparatus for the programmable processing of digital information including the transfer of digital information with memory apparatus and with peripheral apparatus by way of any of first and second duplicative buses, the improvement comprising

A. first and second programmable digital data processing means, each processing means being arranged for receiving and for producing information-transferring signals and responding identically as the other to identical received signals, and being arranged for applying produced signals to at least one said bus,

B. multiplex means connected with both said processing means for applying information-transferring signals from any one of said first and second buses identically to both said processing means,

C. means for comparing produced signals from said first processing means with those from said second processing means and for producing a fault-reporting signal in response to a selected difference between said compared signals, and

D. means responsive to a fault-reporting signal for disabling both said processing means from applying signals to either of said first and second buses.

2. In apparatus according to claim 1, the further improvement wherein said first and second processing means apply identical produced signals to both of said first and second buses independent of the bus from which said multiplexor means applies signals to said processing means.

3. In apparatus according to claim 1, the further improvement wherein said first and second processing means apply identical produced signals to said first and second buses in synchronism with each other.

4. In apparatus according to claim 1, the further improvement in which said first and second processing means normally produce identical signals and apply said identical produced signals in lock-step synchronism with one another to said first and second buses, for all signal applying conditions of said multiplexor means.

5. In apparatus according to claim 4, the further improvement comprising timing control means connected with both said processing means and with said comparing means for applying said produced signals to said buses concurrent with the comparison of produced signals from said first processing means with corresponding produced signals from said second processing means.

6. In apparatus according to claim 1, the further improvement comprising

means for comparing at least selected signals received from said first bus with corresponding signals received from said second bus, and producing a fault-reporting signal in response to a selected difference between said compared signals.

7. In apparatus according to claim 1 for connection to a third bus in addition to said first and second buses, the further improvement comprising

control means arranged for receiving signals from said third bus and for applying signals to said third bus, said control means producing operating signals and applying them identically to said first and second processing means.

8. In apparatus according to claim 7, the further improvement in which said control means includes means for producing signals for controlling the signal-applying operation of said multiplexor means.

9. In apparatus according to claim 7, the further improvement in which said control means includes means connected with said comparing means and responding to said fault-reporting signal for applying a bus-error signal to at least one of said third bus and each of said first and second buses.

10. In apparatus according to claim 7, the further improvement comprising timing control means connected with said first and second processing means for applying signals to at least one of said first and second buses concurrent with said comparison of produced signals from said first processing means with produced signals from said second processing means, and for applying a fault-reporting signal in response to said comparison to at least one of said third bus and each of said first and second buses.

11. In apparatus according to claim 1, the further improvement comprising virtual memory addressing means arranged for receiving virtual address signals from each said processing means and for producing a single set of physical address signals in response thereto.

12. In apparatus according to claim 11, the further improvement comprising means for testing the parity of said physical address signals from said virtual memory addressing means, and for producing a fault-reporting signal in response to a parity error.

13. In central processing apparatus for connection with a bus structure having first and second duplicative buses and a third bus, and for programmable processing of digital information including the transfer of digital information with computer memory apparatus and with computer peripheral apparatus by way of that bus structure, the improvement comprising

A. first and second programmable digital data processing means, eash processing means being arranged for receiving signals, and for producing identical signals as the other one in response to identical received signals, and being arranged for applying produced signals to a different one of said first and second buses,

B. multiplex means connected with said processing means and responding to select signals for applying signals from either of said first and second buses identically to both said processing means,

C. control means for receiving signals from said third bus and for applying operating signals produced in response thereto to both said processing means, and

D. comparator means arranged for comparing produced signals from said first processing means with produced signals from said second processing means, and for producing a fault-reporting signal in response to an invalid comparison.

14. In apparatus according to claim 13, the further improvement comprising means responsive to a fault-reporting signal for disabling the application of produced signals to said first and second buses.

15. In apparatus according to claim 13, the further improvement comprising timing control means for applying produced signals from said two processing means synchronously to both said first and second buses, and for operating said comparing means concurrently to compare said produced signals.

16. In apparatus according to claim 15, the further improvement comprising means in said control means for responding to said fault-reporting signal to apply an error-signal to said third bus.

17. In central processing apparatus having programmable digital data processing means, said apparatus transferring digital signals with memory apparatus and with peripheral apparatus by way of a bus structure having first and second duplicative buses and a third bus, the improvement comprising

A. first and second means for connection to each of first and second duplicative buses for the transfer of signals between the first and second buses and said processing means,

B. third means for connection to a third bus for the transfer of signals between the third bus and said processing means,

C. input selection means responsive to a selection signal for effecting response by said processing means only to signals received on any one of the first and second buses connected with said first and second connection means,

D. output means for applying output signals from said processing means identically to both of the first and second buses connected to said first and second connection means,

E. fault detecting means for detecting a fault condition in the output signals from said processing means and for disabling the application of such output signals to any of said first and second connection means,

F. power supply means for providing electrical power to at least said processing means, said selection means and said output means, and

G. power-responsive means for responding to failing electrical power from said supply means for disabling said output means from applying signals to at least said first and second connection means.

18. In the processing of digital information in programmable central processing apparatus, including the transfer of digital information with memory apparatus and with peripheral apparatus by way of any of first and second duplicative buses, the improvement comprising the steps of

A. applying signals from any of said first and second buses identically to first and second programmable digital data processing means,

B. producing with each of said first and second processing means identical signals in response to said identical applied signals for application to any of said first and second buses,

C. comparing said produced signals from said first processing means with said produced signals from said second processing means and producing a fault reporting signal in response to a selected difference between said compared signals, and

D. disabling the application of produced signals from either of said processing means to either of said first and second buses in response to said fault reporting signal.

19. In information processing according to claim 18, the further improvement comprising the step of operating said first and second processing means in lock-step synchronism for both said signal-applying step and said signal-producing step, and

applying said produced signals to said first and second buses concurrent with said comparing of produced signals.

20. In information processing according to claim 18, the further improvement comprsing the step of applying signals to said first and second processing means from only a single one of said first and second buses and applying said produced signals from said processing means in synchronism to both said first and second buses.

21. In information processing according to claim 18, the further improvement comprising

A. transferring control signals between control means and a third bus separate from said first and second buses and producing with said control means operating signals applied identically to said first and second processing means, and

B. producing with said control means a bus error signal in response to said fault reporting signal and applying said bus error signal to at least one of said third bus and each of said first and second buses.