EP0216170A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Storage management method and system in a multiprocessor environment
Assignee: International Business Machines Corp
•
Filing: 1986-08-26
•
Granted: —
•
Expires: 2006-08-26
Abstract
Abstract A system and method for managing a plurality of mass storage volumes associated with a multiprocessor computer system. Control is allocated to nodes (200, 210, 220, 230) which communicate through a commonly accessible mass storage device (100). A plurality of worker nodes (200,210,220) recognize needs for performance of certain storage mana- gment functions.and communicate those needs to a storage node (230) through transmission of signals to the commonly accessible storage medium. The storage node then retrieves these signals from the commonly accessible storage medium and performs the needed functions. Notification of completion may be communicated to the recognizing worker node by transmission of signals from the storage node to the commonly accessible storage medium. Recognizing worker nodes then retrieve these notification signals to complete the communication path.
Inventors
- 👤 Barbara Jenkins Camacho
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-24 19:28
CPC Classifications
G; G06; G06F; G06F12/00; G06F12/02; G06F12/08; G06F15/00; G06F15/16; G06F15/161; G06F15/163; G06F15/173
Claims Analysis
Total Claims: 27
Claim 1
Independent Claim
126 words
1 . A method of storage management in a multiprocessor environment having at least a commonly accessible storage medium (100) which is accessible to all of the processors (200,210,220,230) in the multiprocessor system, comprising the steps of a) designating one of the processors (230) in the multiprocessor system as a storage node, b) designating the remaining processors (210,220,230) of the multiprocessor system as worker nodes, c) recognizing a service request originated by a process in a worker node; d) transmitting the service request to the commonly accessible storage medium, and e) retrieving the service request from the commonly accessible storage medium by the storage node and performing the requested service, whereby each of the worker nodes and the storage node communicate only through the storage medium.
Claim 2
Dependent Claim
126 words
1 . A method of storage management in a multiprocessor environment having at least a commonly accessible storage medium (100) which is accessible to all of the processors (200,210,220,230) in the multiprocessor system, comprising the steps of a) designating one of the processors (230) in the multiprocessor system as a storage node, b) designating the remaining processors (210,220,230) of the multiprocessor system as worker nodes, c) recognizing a service request originated by a process in a worker node; d) transmitting the service request to the commonly accessible storage medium, and e) retrieving the service request from the commonly accessible storage medium by the storage node and performing the requested service, whereby each of the worker nodes and the storage node communicate only through the storage medium.
Claim 3
Dependent Claim
35 words
The method of Claim 1 including the additional step of transmitting a notification of the performance of the request by the storage node to the storage medium and retrieving the notification by'the recognizing worker node.
Claim 4
Dependent Claim
33 words
The method of Claim 1 in which the step of recognizing a service request includes the step of generating a service request comprising a communication data packet in response to the service request.
Claim 5
Dependent Claim
33 words
The method of Claim 3 in which the step of recognizing a service request includes the step of generating a service request comprising a management work element in response to the service request.
Claim 6
Dependent Claim
49 words
The method of Claim 3 in which the step of transmitting a request to the storage medium includes the step of enqueuing the communication data packet on a storage node queue in a queue data structure on the storage medium, and adjusting a pointer within the communication data packet.
Claim 7
Dependent Claim
67 words
The method of Claim 2 and 5 in which the step of transmitting a notification of the performance of the requested service includes enqueuing the communication data packet in a worker node queue corresponding to the recognizing worker node, in the queue data structure on the storage medium, and adjusting a pointer in a worker node pointer field in a control record of the queue data set.
Claim 8
Dependent Claim
57 words
The method of Claim 5 in which the step of enqueuing the communication data packet on the worker node queue includes the steps of inserting the communication data packet into an available record in the queue data structure and adjusting a pointer in the storage node pointer field in a control record of the queue data structure.
Claim 9
Dependent Claim
35 words
The method of Claim 3 wherein the step of retrieving the service request comprises the steps of: locating the communication data packet and reading the communication data packet into memory associated with the storage node.
Claim 10
Dependent Claim
28 words
The method of Claim 3 in which the step of retrieving the service request includes the step of dequeuing the communication data packet from the storage node queue.
Claim 11
Dependent Claim
62 words
The method of Claim 1 in which the step of performing the needed service is performed by the storage node, the storage node having attached function subtasks, in which the step of performing the requested service includes the step of dispatching the request to a preselected function subtask, and in which the step of requesting includes requesting backup, migration, recall, or restore.
Claim 12
Dependent Claim
169 words
The method of Claim 1 comprising the steps of: a) designating one of processors as a storage node, the storage node having exclusive access to at least one storage volume in addition to access to the commonly accessible storage volume; b) designating the remaining processors as worker nodes, each of the worker nodes having access to at least the commonly accessible storage volume; c) recognizing a need for performance of a storage management function by a worker node; d) transmitting a first signal representative of the required storage management function from the recognizing worker node to the commonly accessible storage volume; e) retrieving the first signal from the commonly accessible storage volume by the storage node; f ) performing the needed storage management function by the storage node; g) transmitting a second signal representative of completion of the needed storage management function from the storage node to the commonly accessible storage volume; h ) retrieving the second signal from the commonly accessible storage volume by the recognizing worker node.
Claim 13
Dependent Claim
48 words
12 . The method of Claim 11 wherein the first signal is transmitted from the recognizing worker node to a predetermined area of the commonly accessible storage volume and the second signal is transmitted from the storage node to that predetermined area of the commonly accessible storage volume.
Claim 14
Dependent Claim
26 words
The method of Claim 11 wherein the second signal is transmitted from the storage node to a second predefined area of the commonly accessible storage volume.
Claim 15
Dependent Claim
52 words
The method of Claim 11 wherein the storage volumes contain a plurality of data sets, each data set having a predetermined amount of space allocated to it, and wherein the step of performing the storage management function includes determining the allocation of storage space on the storage volumes to the data sets.
Claim 16
Dependent Claim
108 words
The method of Claim 14 wherein the allocation-determining step includes copying of a data set on a storage volume to non-volatile storage media, copying of a data set from non-volative storage media to one of the storage volumes, copying a data set from a.first storage volume which is a primary volume, to a second storage volume which is a secondary volume, and subsequently erasing the data set from the first storage volume, or copying a data set from a second storage volume which is a secondary volume, to a first storage volume which is a primary volume, and subsequently erasing the data set from the second storage volume.
Claim 17
Dependent Claim
65 words
The method of Claim 14 wherein the allocation-determining step includes the storage node maintaining a record of space allocation to a plurality of data sets, the step of maintaining a record of space allocation includes storing the maintained record as a data set, and wherein the storage step includes storage of the record on the storage volume which is exclusively accessible to the storage node.
Claim 18
Dependent Claim
181 words
The method of claim 1 including the additional steps of: assigning a plurality of peripheral data storage volumes to the storage node for storing data for any of the processors; establishing and maintaining a common work queue including a work queue in a one of the peripheral data storage devices the common work queue including a time stamp, and an indicator of storage node operability, the queue including wait type requests and no-wait type requests; and operating the storage node to maintain the time stamp when it is operating with the assigned data storage volumes; when each of the worker nodes is accessing the common work queue to insert a request for access to data storage in the assigned peripheral data storage volumes, operating each worker node to examine the time stamp and the storage node operability indicator; if the examination shows a predetermined elapsed time and storage node operability, then inserting the request into the work queue only if it is a wait type request, otherwise inserting the request into the common work queue if the storage node is operable.
Claim 19
Dependent Claim
40 words
18 . The method of Claim 17 in which the step of establishing a common work queue includes the step of establishing a queue data set including the common work queue and a worker node queue for each worker node.
Claim 20
Dependent Claim
36 words
The method of Claim 17 in which the step of inserting a request for data storage includes enqueuing a request in the common work queue by adjusting a first queue pointer in the common work queue.
Claim 21
Dependent Claim
37 words
2 0. The method of Claim 17 further including the steps of performing the request and inserting the performed service request into a worker node queue by adjusting a second queue pointer in the common work queue.
Claim 22
Dependent Claim
40 words
The method of Claim 1 including the step of establishing a queue data structure, the queue data structure including a plurality of pointers in a control record, a storage node queue and a worker node queue for each worker node.
Claim 23
Independent Claim
115 words
The method of Claims 1 and 21 in which the step of transmitting the request includes enqueuing the request on the storage node queue by adjusting a first pointer in the control record, the step of retrieving the request includes dequeuing the service request by the storage node including adjusting a first pointer, the step of performing the requested service is followed by the step of enqueuing the request in the worker node queue of the originating worker node by adjusting a second pointer in the control record, and in which the step of enqueuing in the worker node queue is followed by the step of notifying the requesting process of completion of the request.
Claim 24
Dependent Claim
116 words
A storage management system in a multiprocessor environment having at least a commonly accessible storage means (100) which is accessible to all of the processors (200,210,220,230) in the multiprocessor system, comprising: a plurality of worker node processor means (20 0 , 210,220,230) for recognizing a service request originated by a process, means for transmitting the'service requests to the commonly accessible storage medium, and storage node processor means (230) having means for retrieving the service requests from the commonly accessible storage means whereby the storage node processor means and the worker node processor means communicate only through the commonly accessible storage means, the storage node processor means including service performance means (60) for performing the service requests.
Claim 25
Dependent Claim
116 words
A storage management system in a multiprocessor environment having at least a commonly accessible storage means (100) which is accessible to all of the processors (200,210,220,230) in the multiprocessor system, comprising: a plurality of worker node processor means (20 0 , 210,220,230) for recognizing a service request originated by a process, means for transmitting the'service requests to the commonly accessible storage medium, and storage node processor means (230) having means for retrieving the service requests from the commonly accessible storage means whereby the storage node processor means and the worker node processor means communicate only through the commonly accessible storage means, the storage node processor means including service performance means (60) for performing the service requests.
Claim 26
Dependent Claim
25 words
The system of Claim 23 in which the storage node processor means includes means for notifying the originating process of completion of the service request.
Claim 27
Dependent Claim
42 words
The system of Claim 23 wherein the storage node processor means includes means for transmitting a notification of the performance of the service request to the commonly accessible storage means, and the worker node processor means including means for retrieving the notification.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.