Composite computer system

• US 5,974,565 A
• Filed: 12/18/1996
• Issued: 10/26/1999
• Est. Priority Date: 12/20/1995
• Status: Expired due to Term

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Maeda, Shingo, Morishima, Kikuko, Morishima, Hiroshi, Okuhara, Susumu
• Primary Examiner(s): ELISCA, PIERRE E
• Application Number:
  US08/768,969
• Time in Patent Office: 1,042 Days
• Field of Search: 395/182.02, 395/182.09, 395/182.08, 395/184.01, 395/185.01, 395/185.1, 395/183.13, 395/183.19, 395/183.22, 395/200.54, 395/200.5 , 395/200.51, 395/200.53, 714/11, 714/10, 714/4, 714/47, 714/48, 714/57, 714/46
• US Class Current: 714/11
• CPC Class Codes: G06F 11/16 Error detection or correction of the data by redundancy in hardware
• CPC Class Codes: G06F 11/2007 using redundant communication media

Abstract

A composite computer system which includes a plurality of processors connected to each other by a communication apparatus, an operation supervisory unit for recording operating conditions of the processors when the processors are activated or deactivated, an operation supervisory network for connecting the processors and the operation supervisory unit and a program condition management apparatus for recording operating conditions of programs when the programs of the processors are activated or deactivated. The operation supervisory unit obtains, for fault location purposes, the operating conditions of the processors recorded in the operation supervisory unit via the operation supervisory network and the operating conditions of the programs recorded in the program condition management apparatus.

Claims

1. A composite computer system for permitting access between a plurality of processors connected to each other by a communication apparatus and exclusively controlling common resources connected to said processors, said composite computer system comprising:

an operation supervisory unit which records operating conditions of said processors when at least one of said processors is activated or deactivated;

an operation supervisory network which connects said processors and operation supervisory unit to each other; and

a program condition management apparatus which records operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated,

wherein if one of said processors is unable to access a resource common to said processors because a fault occurs in another of said processors, the one processor determines a location of the fault by obtaining operating conditions of said processors recorded in said operation supervisory unit and operating conditions of said programs recorded in said program condition management apparatus through said operation supervisory network.

2. A composite computer system according to claim 1, wherein said processors supervise operating conditions with respect to each other by transmitting and receiving particular data between each other via said communication apparatus connecting said processors.

3. A composite computer system for permitting access between a plurality of processors connected to each other by a communication apparatus and exclusively controlling common resources connected to said processors, said composite computer system comprising:

an operation supervisory unit which records operating conditions of said processors when at least one of said processors is activated or deactivated;

an operation supervisory network which connects said processors and operation supervisory unit to each other;

a program condition management apparatus which records operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated,

wherein if a fault is generated in one of said processors, a location of the fault is determined by obtaining operating conditions of said processors recorded in said operation supervisory unit and operating conditions of said programs recorded in said program condition management apparatus through said operation supervisory network;

a plurality of said operation supervisory units;

a communication control apparatus which controls communication to said processors from said operation supervisory units; and

an operation supervisory unit multiplexer which multiplexes communication to said operation supervisory units from said processors and controls said communication control apparatus,

wherein operating conditions of said processors are supervised by each of said operation supervisory units by transmitting operating conditions of said processors to others of said operation supervisory units by said operation supervisory unit multiplexer and controlling communication to said processors from said operation supervisory units other than said operation supervisory unit by said communication control apparatus, and, if a fault is generated in said operation supervisory unit, a communication control condition of the communication control apparatus of said operation supervisory unit having generated the fault is cancelled by said operation supervisory unit multiplexer and operating conditions of said processors are supervised continuously by said operation supervisory unit for which the communication condition has been cancelled.

4. A composite computer system according to claim 3, wherein said processors supervise operating conditions with respect to each other by transmitting and receiving particular data between each other via said communication apparatus connecting said processors.

5. A method, in a composite computer system, of permitting access between a plurality of processors connected to each other by communication apparatus and exclusively controlling common resources connected to said processors, said method comprising the steps of:

recording operating conditions of said processors when at least one of said processors is activated or deactivated;

recording operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated; and

if one of said processors is unable to access a resource common to said processors because a fault occurs in another of said processors, determining a location of said fault by obtaining the recorded operating conditions of said processors and recorded operating conditions of said programs.

6. A method according to claim 5, wherein said step of supervising operating conditions of said processors comprises the step of:

transmitting and receiving particular data between said processors via said communication apparatus connecting said processors.

7. A method, in a composite computer system, of permitting access between a plurality of processors connected to each other by a communication apparatus and exclusively controlling common resources connected to said processors, said method comprising the steps of:

recording operating conditions of said processors when at least one of said processors is activated or deactivated;

recording operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated; and

if a fault is generated in one of said processors, determining a location of said fault by obtaining the recorded operating conditions of said processors and recorded operating conditions of said programs;

recording operating conditions of said processors in a plurality of said operation supervisory units;

multiplexing communication to said operation supervisory units from said processors;

supervising operating conditions of said processors by each of said operation supervisory units by transmitting operating conditions of said processors to others of said operation supervisory units and controlling communication to said processors from said operation supervisory units other than said operation supervisory unit; and

if a fault is generated in said operation supervisory unit, cancelling a communication control condition of said operation supervisory unit having generated the fault and supervising operating conditions of said processors continuously by said operation supervisory unit for which the communication condition has been cancelled.

8. A method according to claim 7, wherein said step of supervising operating conditions of said processors comprises the step of:

transmitting and receiving particular data between said processors via said communication apparatus connecting said processors.

9. A computer program stored on a storage medium executable by a composite computer system so as to permit access between a plurality of processors connected to each other by a communication apparatus and exclusive control of common resources connected to said processors, said computer program, when executed by said composite computer system, causes said composite computer system to perform the steps of:

recording operating conditions of said processors when at least one of said processors is activated or deactivated;

recording operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated; and

if one of said processors is unable to access a resource common to said processors because a fault occurs in another of said processors, determining a location of said fault by obtaining the recorded operating conditions of said processors and recorded operating conditions of said programs.

10. A computer program according to claim 9, wherein said step of supervising operating conditions of said processors comprises the step of:

transmitting and receiving particular data between said processors via said communication apparatus connecting said processors.

11. A computer program stored on a storage medium executable by a composite computer system so as to permit access between a plurality of processors connected to each other by a communication apparatus and exclusive control of common resources connected to said processors, wherein said computer program, when executed by said composite computer system, causes said composite computer system to perform the steps of:

recording operating conditions of said processors when at least one of said processors is activated or deactivated;

recording operating conditions of a plurality of programs being executed by said processors when at least one program is activated or deactivated;

if a fault is generated in one of said processors, determining a location of said fault by obtaining the recorded operating conditions of said processors and recorded operating conditions of said programs;

recording operating conditions of said processors in a plurality of operation supervisory units;

multiplexing communication to said operation supervisory units from said processors;

supervising operating conditions of said processors by each of said operation supervisory units by transmitting operating conditions of said processors to others of said operation supervisory units and controlling communication to said processors from said operation supervisory units other than said operation supervisory unit; and

if a fault is generated in said operation supervisory unit, cancelling a communication control condition of said operation supervisory unit having generated the fault and supervising operating conditions of said processors continuously by said operation supervisory unit for which the communication condition has been cancelled.

12. A computer program according to claim 11, wherein said step of supervising operating conditions of said processors comprises the step of:

transmitting and receiving particular data between said processors via said communication apparatus connecting said processors.

13. An operation supervisory apparatus for use in a composite computer system having a plurality of processors connected to each other by a communication apparatus wherein said operation supervisory unit permits access between the processors and controls common resources connected to the processors, said operation supervisory apparatus comprising:

a processor communication apparatus which controls communication between said processors and said operation supervisory apparatus;

an operating condition management apparatus which records operating conditions of said processors when said processors are activated or deactivated;

means for determining whether a communication from a communicating processor is an inquiry, activation notice or deactivation notice;

means for retrieving operating condition information of another processor and notifying said operating condition information to the communicating processor when said communication is an inquiry;

means for changing an operating condition of said another processor from a deactivation state to an activation state and notifying the communicating processor that said operating condition has been changed when said communication is an activation notice; and

means for changing an operating condition of said another processor from an activation state to a deactivation state and notifying the communicating processor that a deactivation has occurred when said communication is a deactivation notice.

14. An operation supervisory apparatus according to claim 13, wherein each processor comprises:

an exclusive control management program, being executed by said processor for supervising operating conditions of processors other than said processor by use of said operation supervisory unit.

15. An operation supervisory apparatus for use in a composite computer system having a plurality of processors connected to each other by a communication apparatus wherein said operation supervisory unit permits access between the processors and controls common resources connected to the processors, said operation supervisory apparatus comprising:

a processor communication apparatus which controls communication between said processors and said operation supervisory apparatus,

wherein each processor comprises an operating system which controls operation of said processor, wherein said operating system includes a program condition management apparatus for storing operating conditions of a computer program being executed by said processor;

an operating condition management apparatus which records operating conditions of said processors when said processors are activated or deactivated;

means for determining whether a communication from a communicating processor is an inquiry, activation notice or deactivation notice;

means for retrieving operating condition information of another processor and notifying said operating condition information to the communicating processor when said communication is an inquiry;

means for changing an operating condition of said another processor from a deactivation state to an activation state and notifying the communicating processor that said operating condition has been changed when said communication is an activation notice; and

means for changing an operating condition of said another Processor from an activation state to a deactivation state and notifying the communicating processor that a deactivation has occurred when said communication is a deactivation notice.

16. An operation supervisory apparatus according to claim 15, wherein each processor comprises:

an exclusive control management program, being executed by said processor, for controlling operating conditions of processors other than said processor by use of said operation supervisory unit.

17. A method, in an operation supervisory apparatus included in a composite computer system having a plurality of processors connected to each other by a communication apparatus, of permitting access between the processors and controlling common resources connected to the processors, said method comprising the steps of:

controlling communication between said processors and said operation supervisory apparatus;

recording operating conditions of said processors when said processors are activated or deactivated;

determining whether a communication from a communicating processor is an inquiry, activation notice or deactivation notice;

retrieving operating condition information of another processor and notifying said operating condition information to the communicating processor when said communication is an inquiry;

changing an operating condition of said another processor from a deactivation state to an activation state and notifying the communicating processor that said operation condition has been changed when said communication is an activation notice; and

changing an operation condition of said another processor from an activation state to a deactivation state and notifying the communicating processor that a deactivation has occurred when said communication is a deactivation notice.

18. A method according to claim 17, wherein each processor comprises:

an operating system which controls operation of said processor, wherein said operating system includes a program condition management apparatus for storing operating conditions of a computer program being executed by said processor.

19. A method according to claim 17, wherein each processor comprises:

an exclusive control management program, being executed by said processor for supervising operating conditions of processors other than said processor by use of said operation supervisory unit.

20. A method according to claim 18, wherein each processor comprises:

an exclusive control management program, being executed by said processor, for controlling operating conditions of processors other than said processor by use of said operation supervisory unit.

21. A computer program stored on a storage medium for performing operation supervisory functions when executed by a composite computer system having a plurality of processors connected to each other by a communication apparatus so as to permit access between the processors and control common resources connected to the processors, said computer program, when execute by said composite system, causes said composite computer system to perform the steps of:

controlling communication between said processors and an operation supervisory apparatus;

recording operating conditions of said processors when said processors are activated or deactivated;

determining whether a communication from a communicating processor is an inquiry, activation notice or deactivation notice;

retrieving operating condition information of another processor and notifying said operating condition information to the communicating processor when said communication is an inquiry;

changing an operating condition of said another processor from a deactivation state to an activation state and notifying the communicating processor that said operating condition has been changed when said communication is an activation notice; and

changing an operating condition of said another processor from an activation state to a deactivation state and notifying the communicating processor that a deactivation has occurred when said communication is a deactivation notice.

22. A computer program according to claim 21, wherein each processor comprises:

an operating system which controls operation of said processor, wherein said operating system includes a program condition management apparatus for storing operating conditions of a computer program being executed by said processor.

23. A computer program according to claim 21, wherein each processor comprises:

an exclusive control management program, being executed by said processor for supervising operating conditions of processors other than said processor by use of said operation supervisory unit.

24. A computer program according to claim 22, wherein each processor comprises:

an exclusive control management program, being executed by said processor, for controlling operating conditions of processors other than said processor by use of said operation supervisory unit.