IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0748579
(2010-03-29)
|
등록번호 |
US-8606979
(2013-12-10)
|
발명자
/ 주소 |
- Archer, Charles J.
- Blocksome, Michael A.
- Ratterman, Joseph D.
- Smith, Brian E.
|
출원인 / 주소 |
- International Business Machines Corporation
|
대리인 / 주소 |
Biggers Kennedy Lenart Spraggins, LLP
|
인용정보 |
피인용 횟수 :
0 인용 특허 :
38 |
초록
▼
Distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network including assigning the root node of the operational group to send acknowledgments for lock requests, the root lock administration module comprising a module of automated computin
Distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network including assigning the root node of the operational group to send acknowledgments for lock requests, the root lock administration module comprising a module of automated computing machinery; receiving a lock request assigned to a particular node from a child node; determining whether another request from another child is directly ahead in an acknowledgement queue; if a request from another child is directly ahead in the acknowledgement queue, putting the lock request for the particular node in the acknowledgement queue until the lock request directly ahead in the acknowledgement queue is satisfied and when the lock request ahead in the queue is satisfied, sending the particular node for whom the lock request is assigned a message acknowledging the particular node has the lock; and if a request from another child is not directly ahead in a queue, sending to the particular node for whom the lock request is assigned a message acknowledging that the particular node has the lock.
대표청구항
▼
1. A computer implemented method of distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network, the method comprising: assigning, by a root lock administration module, the root node of the operational group to send acknowledgments for loc
1. A computer implemented method of distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network, the method comprising: assigning, by a root lock administration module, the root node of the operational group to send acknowledgments for lock requests, the root lock administration module comprising a module of automated computing machinery;receiving, by a root lock administration module, a lock request for a particular node from a child node;determining, by the root lock administration module, whether another request from another child is directly ahead in an acknowledgement queue;if a request from another child is directly ahead in the acknowledgement queue, putting, by the root lock administration module, the lock request for the particular node in the acknowledgement queue until the lock request directly ahead in the acknowledgement queue is satisfied and when the lock request ahead in the queue is satisfied, sending the particular node for whom the lock request is for a message acknowledging the particular node has the lock; andif a request from another child is not directly ahead in a queue, sending, by the root lock administration module, to the particular node for whom the lock request is for a message acknowledging that the particular node has the lock. 2. The method of claim 1 further comprising for all parent nodes other than the root node: receiving, by a parent lock administration module, a lock request for a particular node from a child node, the parent lock administration module comprising a module of automated computing machinery;determining, by the parent lock administration module, whether another request from another child is directly ahead in a queue;if a request from another child node directly is ahead in a queue, putting, by a parent lock administration module, the lock request for the particular node in the queue until the lock request from another child directly ahead in the queue is satisfied and when the lock request directly ahead in the queue is satisfied, sending, by a parent lock administration module, the lock request to a grandparent node above the parent node in the operational group; andif a request from another child is not directly ahead in a queue, sending, by a parent lock administration module, the lock request to the grandparent node above the parent above in the operational group. 3. The method of claim 1 further comprising sending, by the parent lock administration module for the particular node with the lock, a lock release to a parent node above the particular node in the operation group when the particular node has completed work requiring the lock. 4. The method of claim 1 further comprising for all parent nodes other than the root: receiving, by the parent lock administration module, a lock release for a particular node from a child node;sending, by the parent lock administration module, another lock release for the same particular node to the grandparent above parent node in the operational group; andsending, by the parent lock administration module, a lock request for another particular node to the grandparent above in the parent node in the operational group for the next request in the queue. 5. The method of claim 1 further comprising for the root: receiving, by the root lock administration module, a lock release from a child node;selecting, by the root lock administration module, the next lock request in the acknowledgement queue for a next particular node; andsending, by the root lock administration module, to the next particular node for whom the next lock request is for a message acknowledging the next particular node has the lock. 6. The method of claim 1 wherein in a hierarchical tree structured network is implemented in a distributed processing system that further comprises a parallel computer that includes: a plurality of compute nodes;a first data communications network coupling the compute nodes for data communications and optimized for point to point data communications; anda second data communications network that includes data communications links coupling the compute nodes so as to organize the compute nodes as a tree, each compute node having a separate arithmetic logic unit (‘ALU’) dedicated to parallel operations. 7. The method of claim 1 wherein in a hierarchical tree structured network is implemented in a distributed processing system that further comprises a hybrid computing environment, the hybrid computing environment comprising: a host computer having a host computer architecture; andan accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module. 8. An apparatus for distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network, the apparatus comprising a computer processor and a computer memory operatively coupled to the computer processor, the computer memory having disposed within it computer program instructions for: assigning, by a root lock administration module, the root node of the operational group to send acknowledgments for lock requests, the root lock administration module comprising a module of automated computing machinery;receiving, by a root lock administration module, a lock request for to a particular node from a child node;determining, by the root lock administration module, whether another request from another child is directly ahead in an acknowledgement queue;if a request from another child is directly ahead in the acknowledgement queue, putting, by the root lock administration module, the lock request for the particular node in the acknowledgement queue until the lock request directly ahead in the acknowledgement queue is satisfied and when the lock request ahead in the queue is satisfied, sending the particular node for whom the lock request is for a message acknowledging the particular node has the lock; andif a request from another child is not directly ahead in a queue, sending, by the root lock administration module, to the particular node for whom the lock request is for a message acknowledging that the particular node has the lock. 9. The apparatus of claim 8 wherein the computer memory also has disposed within it computer program instructions for: receiving, by a parent lock administration module, a lock request for a particular node from a child node, the parent lock administration module comprising a module of automated computing machinery;determining, by the parent lock administration module, whether another request from another child is directly ahead in a queue;if a request from another child node directly is ahead in a queue, putting, by a parent lock administration module, the lock request for the particular node in the queue until the lock request from another child directly ahead in the queue is satisfied and when the lock request directly ahead in the queue is satisfied, sending, by a parent lock administration module, the lock request to a grandparent node above the parent node in the operational group; andif a request from another child is not directly ahead in a queue, sending, by a parent lock administration module, the lock request to the grandparent node above the parent above in the operational group. 10. The apparatus of claim 8 wherein the computer memory also has disposed within it computer program instructions for sending, by the parent lock administration module for the particular node with the lock, a lock release to a parent node above the particular node in the operation group when the particular node has completed work requiring the lock. 11. The apparatus of claim 8 wherein the computer memory also has disposed within it computer program instructions for: receiving, by the parent lock administration module, a lock release for a particular node from a child node;sending, by the parent lock administration module, another lock release for the same particular node to the grandparent above parent node in the operational group; andsending, by the parent lock administration module, a lock request for another particular node to the grandparent above in the parent node in the operational group for the next request in the queue. 12. The apparatus of claim 8 wherein the computer memory also has disposed within it computer program instructions for: receiving, by the root lock administration module, a lock release from a child node;selecting, by the root lock administration module, the next lock request in the acknowledgement queue for a next particular node; andsending, by the root lock administration module, to the next particular node for whom the next lock request is for a message acknowledging the next particular node has the lock. 13. A computer program product for distributed administration of a lock for an operational group of compute nodes in a hierarchical tree structured network, the computer program product disposed in a computer readable storage medium, wherein the computer readable storage medium is not a signal and the computer program product comprises computer program instructions for: assigning, by a root lock administration module, the root node of the operational group to send acknowledgments for lock requests, the root lock administration module comprising a module of automated computing machinery;receiving, by a root lock administration module, a lock request for to a particular node from a child node;determining, by the root lock administration module, whether another request from another child is directly ahead in an acknowledgement queue;if a request from another child is directly ahead in the acknowledgement queue, putting, by the root lock administration module, the lock request for the particular node in the acknowledgement queue until the lock request directly ahead in the acknowledgement queue is satisfied and when the lock request ahead in the queue is satisfied, sending the particular node for whom the lock request is for a message acknowledging the particular node has the lock; andif a request from another child is not directly ahead in a queue, sending, by the root lock administration module, to the particular node for whom the lock request is for a message acknowledging that the particular node has the lock. 14. The computer program product of claim 13 further comprising computer program instructions for: receiving, by a parent lock administration module, a lock request for a particular node from a child node, the parent lock administration module comprising a module of automated computing machinery;determining, by the parent lock administration module, whether another request from another child is directly ahead in a queue;if a request from another child node directly is ahead in a queue, putting, by a parent lock administration module, the lock request for the particular node in the queue until the lock request from another child directly ahead in the queue is satisfied and when the lock request directly ahead in the queue is satisfied, sending, by a parent lock administration module, the lock request to a grandparent node above the parent node in the operational group; andif a request from another child is not directly ahead in a queue, sending, by a parent lock administration module, the lock request to the grandparent node above the parent above in the operational group. 15. The computer program product of claim 13 further comprising computer program instructions for sending, by the parent lock administration module for the particular node with the lock, a lock release to a parent node above the particular node in the operation group when the particular node has completed work requiring the lock. 16. The computer program product of claim 13 further comprising computer program instructions for: receiving, by the parent lock administration module, a lock release for a particular node from a child node;sending, by the parent lock administration module, another lock release for the same particular node to the grandparent above parent node in the operational group; andsending, by the parent lock administration module, a lock request for another particular node to the grandparent above in the parent node in the operational group for the next request in the queue. 17. The computer program product of claim 13 further comprising computer program instructions for: receiving, by the root lock administration module, a lock release from a child node;selecting, by the root lock administration module, the next lock request in the acknowledgement queue for a next particular node; andsending, by the root lock administration module, to the next particular node for whom the next lock request is for a message acknowledging the next particular node has the lock. 18. The computer program product of claim 13 wherein in a hierarchical tree structured network is implemented in a distributed processing system that further comprises a parallel computer that includes: a plurality of compute nodes;a first data communications network coupling the compute nodes for data communications and optimized for point to point data communications; anda second data communications network that includes data communications links coupling the compute nodes so as to organize the compute nodes as a tree, each compute node having a separate arithmetic logic unit (‘ALU’) dedicated to parallel operations. 19. The computer program product of claim 13 wherein in a hierarchical tree structured network is implemented in a distributed processing system that further comprises a hybrid computing environment, the hybrid computing environment comprising: a host computer having a host computer architecture; andan accelerator having an accelerator architecture, the accelerator architecture optimized, with respect to the host computer architecture, for speed of execution of a particular class of computing functions, the host computer and the accelerator adapted to one another for data communications by a system level message passing module.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.