Abstract: A system for managing locks that give permission to access resources in a computer system is disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. The master resource object includes a convert queue used to maintain a list of convert requests made by the shadow resource objects on the various other nodes. The convert request at the head of the global convert queue is the “head convert request”.

Abstract: A system for managing locks that give permission to access resources in a computer system are disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. Each shadow lock holds lock information that is used to detect deadlocks. The information may be used to generate bipartite directed graphs that include nodes representing resources and nodes representing transactions.

Abstract: A system for managing locks that give permission to access resources in a computer system is disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. According to an aspect of the invention, the master resource object stores data indicating what is knows about a shadow resource object. The master node increments a sequence number when the master resource object changes what is knows about the shadow resource object.

Abstract: A system for managing locks that give permission to access resources in a computer system is disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. The master resource object includes a convert queue used to maintain a list of convert requests made by various processes. The lock request for shadow resource objects are granted when the lock request is at the head of the convert queue.

Abstract: A system for managing locks that give permission to access resources in a computer system is disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. The shadow resource object includes a convert queue used to maintain a list of convert requests made by various processes. The lock request for the process is granted when the lock request is at the head of the convert queue.

Abstract: The lock management system disclosed is distributed i.e. information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire access to the resource. A master resource object grants locks to shadow resource objects located on the nodes. Each shadow resource object grants locks on the resource to processes that are located on the same node as the shadow resource object. Data stored in the same node of a shadow resource object indicates what information the shadow resource object has sent to the master object. The information includes what lock mode is held by a shadow resource object and what lock mode is currently desired by the shadow resource lock. The shadow resource object transmits both information and requests to the master resource object based on the data stored in the same node of a shadow resource object.

Abstract: A system for managing locks that give permission to access resources in a computer system are disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located processes that desire to access to the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. A shadow resource object may retain a lock on a resource even when no process on the same node requires the shadow resource to retain the lock.

Abstract: A method and system for managing locks that give permission to access resources in a computer system are disclosed. The lock management system is distributed and provides deadlock protection. The lock management system is distributed in that information about which locks have been granted for a resource is stored at both a master node and at the nodes on which are located the processes that desired to access the resource. A master resource object located on the master node grants locks to shadow resource objects located on the nodes on which are located the processes that desire to access the resource. Each shadow resource object grants locks on the resource to the processes that are located on the same node as the shadow resource object. The lock owned by the shadow resource object determines the types of locks the shadow resource object is allowed to grant to processes.

Abstract: A method and apparatus for managing access to resources is provided. When a process requires access to a resource, the process requests a lock on the resource from a lock manager unit that resides on the same node as the process. If a resource object for the resource does not exists, one is created in the lock manager unit, but not on lock manager units on other nodes. Because each lock manager unit does not have to store all resource objects, and resource objects are only created for resources that are actually used, the overhead of the lock management system is significantly reduced. Resources are grouped in recovery domains. When a lock manager unit that supported a recovery domain fails, the recovery domain is marked invalid. All resources in the recovery domain are considered invalid unless it would have been impossible for a failed instance to have held an exclusive lock on the resource. A snapshot of lock information is made before cleanup is performed on invalid resources.

Abstract: A method and apparatus for managing access to resources is provided. When a process requires access to a resource, the process requests a lock on the resource from a lock manager unit that resides on the same node as the process. If a resource object for the resource does not exist, one is created in the lock manager unit, but not on lock manager units on other nodes. Because each lock manager unit does not have to store all resource objects, and resource objects are only created for resources that are actually used, the overhead of the lock management system is significantly reduced. Resources are grouped in recovery domains. When a lock manager unit that supported a recovery domain fails, the recovery domain is marked invalid. All resources in the recovery domain are considered invalid unless it would have been impossible for a failed instance to have held an exclusive lock on the resource. A snapshot of lock information is made before cleanup is performed on invalid resources.