Patent US3107303A successfully refreshed from Google Patents.
Patent US3624197A successfully refreshed from Google Patents.
US4480304A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Method and means for the retention of locks across system, subsystem, and communication failures in a multiprocessing, multiprogramming, shared data environment
Assignee: International Business Machines Corp
•
Filing: 1980-10-06
•
Granted: 1984-10-30
•
Expires: 2000-10-06
Abstract
Abstract Method and means for operating a computing system for controlling access to and maintaining the integrity of data resources shared by multiple applications executing on a plurality of central electronic complexes (CEC's) while minimizing communication of sharing control parameters between different electronic complexes. Each complex includes a resource lock manager (IRLM) which maintains the hold and wait locks for applications executing under one or more information management subsystems on the complex and selected wait locks for the other complex. Selective communication of lock request information is controlled by hash tables maintained in synchronization in each IRLM, which denote the interest of each complex in each hash class, or congruence class, of data resources. A first complex already having exclusive use of a congruence class grants locks on data resources which map into the congruence class without communication of the request to the other complex. When such communication is required, and the other complex determines that an inconsistent hold lock already exists for the data resource of the request, the lock request is waited in both IRLM's. When the inconsistant hold lock is subsequently released, the wait lock is deleted from the IRLM of the other complex, and the first complex grants the requested lock. Each IRLM also includes an identified subsystem list (ISL), which is maintained in synchronization and used along with the hash tables and hold/wait locks to retain locks across CEC, subsystem, and communication link failures and to enhance recovery thereafter.
Inventors
- 👤 Clarence A. Carr
- 👤 Robert L. Huddleston
- 👤 Jimmy P. Strickland
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-23 14:21
CPC Classifications
G; G06; G06F; G06F9/00; G06F9/06; G06F9/46; G06F9/52
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
232 words
A method for operating a general purpose computing system including at least two central electronic complexes interconnected by a communication link and sharing access to at least one data base having a plurality of data resources, the method inhibiting access to data resources that may be inconsistent due to failure in a complex or in the communication link, comprising the steps of: responsive to failure of a first complex, operating the second complex to inhibit access by the second complex on behalf of a transaction to data resources in which said first complex had an interest prior to the complex failure by terminating said transaction and backing out any changes to said data base made by said transaction; responsive to failure of the communication link, operating each complex to inhibit access thereby on behalf of a transaction to data resources of a congruence class in which it did not have private use prior to the link failure by terminating said transaction and backing out any changes to said data base made by said transaction; and responsive to failure of a first complex, retaining in the lock control structure of a second complex the congruence classes in which the first complex had an interest prior to the complex failure until completion of restart/recovery processing for all subsystems denoted in the lock control structure of the second complex as identified to the first complex.
Claim 2
Independent Claim
232 words
A method for operating a general purpose computing system including at least two central electronic complexes interconnected by a communication link and sharing access to at least one data base having a plurality of data resources, the method inhibiting access to data resources that may be inconsistent due to failure in a complex or in the communication link, comprising the steps of: responsive to failure of a first complex, operating the second complex to inhibit access by the second complex on behalf of a transaction to data resources in which said first complex had an interest prior to the complex failure by terminating said transaction and backing out any changes to said data base made by said transaction; responsive to failure of the communication link, operating each complex to inhibit access thereby on behalf of a transaction to data resources of a congruence class in which it did not have private use prior to the link failure by terminating said transaction and backing out any changes to said data base made by said transaction; and responsive to failure of a first complex, retaining in the lock control structure of a second complex the congruence classes in which the first complex had an interest prior to the complex failure until completion of restart/recovery processing for all subsystems denoted in the lock control structure of the second complex as identified to the first complex.
Claim 3
Dependent Claim
82 words
The method of claim 1 in which at least one of the central electronic complexes includes a plurality of information management subsystems sharing access to said data base, the method further inhibiting access to data resources that may be inconsistent due to failure in a subsystem, comprising the further step of: responsive to failure of a subsystem, operating each complex to inhibit access by other subsystems to data resources in which the failed subsystem held a lock prior to the subsystem failure.
Claim 4
Dependent Claim
54 words
The method of claim 2, comprising the further step of maintaining in each complex a lock control structure, each lock control structure denoting the congruence classes in which each complex has an interest and those for which each complex has private use, and further denoting the complex subsystems identified to both lock control structures.
Claim 5
Dependent Claim
28 words
The method of claim 3, further comprising the step of operating each complex to maintain within its lock control structure locks on data resources held by its subsystems.
Claim 6
Dependent Claim
43 words
The method of claim 4 further comprising the step responsive to failure of a first subsystem of retaining in its associated lock control structure locks held by said first subsystem until such locks are released by restart or recovery of the first subsystem.
Claim 7
Dependent Claim
45 words
The method of claim 3 further comprising the step responsive to failure of the communication link and subsequent recovery of communications of exchanging between complexes the lock control structure data denoting congruence classes of interest or private use, and the data denoting the identified subsystems.
Claim 8
Independent Claim
280 words
A method for controlling access to shared data resources in a multiprocessing, multiprogramming general purpose computing environment to inhibit access to data resources that may be inconsistent due to failures of a communication link, operating system, subsystem, or lock control structure, the method comprising the steps of: maintaining within each of a plurality of central electronic complexes a lock control structure, the lock control structure including a global hash table containing for each complex an indication of interest in data resources in a plurality of congruence classes, the lock control structure also including a subsystem list identifying each subsystem associated with any lock control structure; responsive to failure of a first subsystem, retaining within the lock control structure all locks previously granted to said first subsystem until released by restart or recovery of said first subsystem, and enabling access by the other subsystems only to specific data resources not locked by said first subsystem; responsive to failure of a first operating system on a first complex, retaining in the lock control structure of the other complexes the global hash table entries for the first complex, inhibiting the granting of new locks for data resources in those congruence classes in which the first complex had an interest until the global hash table entries for the first complex have been cleared by restart/recovery of all subsystems associated with the lock control structure of the first complex; responsive to failure of a communication link interconnecting two complexes, operating each complex to inhibit the granting of locks for any data resource of a congruence class in which the complex did not have a prior interest together with no prior interest by the other complex.
Claim 9
Independent Claim
280 words
A method for controlling access to shared data resources in a multiprocessing, multiprogramming general purpose computing environment to inhibit access to data resources that may be inconsistent due to failures of a communication link, operating system, subsystem, or lock control structure, the method comprising the steps of: maintaining within each of a plurality of central electronic complexes a lock control structure, the lock control structure including a global hash table containing for each complex an indication of interest in data resources in a plurality of congruence classes, the lock control structure also including a subsystem list identifying each subsystem associated with any lock control structure; responsive to failure of a first subsystem, retaining within the lock control structure all locks previously granted to said first subsystem until released by restart or recovery of said first subsystem, and enabling access by the other subsystems only to specific data resources not locked by said first subsystem; responsive to failure of a first operating system on a first complex, retaining in the lock control structure of the other complexes the global hash table entries for the first complex, inhibiting the granting of new locks for data resources in those congruence classes in which the first complex had an interest until the global hash table entries for the first complex have been cleared by restart/recovery of all subsystems associated with the lock control structure of the first complex; responsive to failure of a communication link interconnecting two complexes, operating each complex to inhibit the granting of locks for any data resource of a congruence class in which the complex did not have a prior interest together with no prior interest by the other complex.
Claim 10
Dependent Claim
292 words
A computing system adapted for inhibiting access to shared data resources that may be inconsistent due to a failure, comprising: first and second central electronic complexes; communication means for interconnecting said complexes; data storage means for storing data resources for sharing by said complexes; each said complex including: a plurality of management subsystems for accessing said data resources; a lock manager for selectively granting, waiting, and denying locks for data resources to said subsystems; each said lock manager including: global hash table means for denoting for each congruence class the interest state of each complex; subsystem list means for denoting each subsystem associated with a lock manager; and held lock means for denoting resource locks held by subsystems in the complex of the lock manager; and operating means in each complex for controlling the operation of the computing system for retaining locks for data resources selectively across failures of a complex, of a subsystem, and of the communication means; the operating means being (1) responsive to a failure of a first subsystem for retaining all locks held by said first subsystem until released by restart or recovery thereof, and for granting locks for enabling access by the other subsystems only to data resources not locked by said first subsystem; (2) responsive to failure of a complex for retaining in the other complex the global hash table entries of the failed complex, and for enabling access by the other complex only to data resources of a congruence class in which the failed complex did not have an interest; and (3) responsive to failure of said communication means for enabling access by each complex to data resources in congruence classes in which the complex had a private use state prior to the communication failure.
Claim 11
Dependent Claim
292 words
A computing system adapted for inhibiting access to shared data resources that may be inconsistent due to a failure, comprising: first and second central electronic complexes; communication means for interconnecting said complexes; data storage means for storing data resources for sharing by said complexes; each said complex including: a plurality of management subsystems for accessing said data resources; a lock manager for selectively granting, waiting, and denying locks for data resources to said subsystems; each said lock manager including: global hash table means for denoting for each congruence class the interest state of each complex; subsystem list means for denoting each subsystem associated with a lock manager; and held lock means for denoting resource locks held by subsystems in the complex of the lock manager; and operating means in each complex for controlling the operation of the computing system for retaining locks for data resources selectively across failures of a complex, of a subsystem, and of the communication means; the operating means being (1) responsive to a failure of a first subsystem for retaining all locks held by said first subsystem until released by restart or recovery thereof, and for granting locks for enabling access by the other subsystems only to data resources not locked by said first subsystem; (2) responsive to failure of a complex for retaining in the other complex the global hash table entries of the failed complex, and for enabling access by the other complex only to data resources of a congruence class in which the failed complex did not have an interest; and (3) responsive to failure of said communication means for enabling access by each complex to data resources in congruence classes in which the complex had a private use state prior to the communication failure.
References Cited (Backward Citations)
US3400371A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1987-10-30 | 1988-04-30 | 1988-10-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1991-10-30 | 1992-04-30 | 1992-10-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1995-10-30 | 1996-04-30 | 1996-10-30 | $7700.00 | Window Closed / Lapsed |