Patent US2164960A successfully refreshed from Google Patents.
EP0266586A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Decreasing response time to I/O request by duplicating data
Assignee: International Business Machines Corp
•
Filing: 1987-10-13
•
Granted: —
•
Expires: 2007-10-13
Abstract
Abstract This invention relates to a method of reducing response time to I/O requests to access data stored in a plurality of circulating storages devices. More specifically, data is duplicated on the circulating storage devices. The access time of accessing an originally stored data unit and a duplicate of that data unit is estimated and the data unit requiring the lesser access time to be accessed is then accessed. This invention also describes the method of writing or storing data in the direct access storage device. More specifically, data is stored through the use of high priority queues and low priority queues. Requests to originally store data or the first copy of data is stored in a high priority queue, while the request for storing the copy of the originally stored data is stored in a low priority queue. Then, all requests in the high priority queue are satisfied, before any request in the low priority queue is satisfied.
Inventors
- 👤 Peter Anthony Franaszek
- 👤 Asser Nasreldin Tantawi
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-28 11:47
CPC Classifications
G; G06; G06F; G06F3/00; G06F3/06; G06F3/0601; G06F3/0602; G06F3/061; G06F3/0611; G06F3/0628; G06F3/0646; G06F3/065; G06F3/0655; G06F3/0659; G06F3/0668; G06F3/0671; G06F3/0683; G06F3/0689; G11; G11B; G11B27/00; G11B27/002
Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
118 words
In a direct access storage system with a plurality of circulating storage devices, a method of decreasing response time to a request to read data stored in the storage devices, comprising the steps of: (a) storing each data unit of a number of data units in each of at least two of the storage devices; (b) estimating the access time for accessing a selected data unit of the data units stored in step (a) in each of the storage devices having the selected data unit stored therein; and (c) accessing the selected data unit in the storage device in which the access time is estimated to be a minimum so that the selected data unit can be read.
Claim 2
Dependent Claim
118 words
In a direct access storage system with a plurality of circulating storage devices, a method of decreasing response time to a request to read data stored in the storage devices, comprising the steps of: (a) storing each data unit of a number of data units in each of at least two of the storage devices; (b) estimating the access time for accessing a selected data unit of the data units stored in step (a) in each of the storage devices having the selected data unit stored therein; and (c) accessing the selected data unit in the storage device in which the access time is estimated to be a minimum so that the selected data unit can be read.
Claim 3
Dependent Claim
21 words
A method as recited in claim 1, wherein each data unit is stored in two of the plurality of storage devices.
Claim 4
Dependent Claim
49 words
A method as recited in claim 1, wherein each storage device is partitioned into a number of partitions, wherein the partitions of data units originally stored are duplicated on a different storage device with no two partitions of originally stored data units being duplicated on the same storage device.
Claim 5
Dependent Claim
27 words
A method as recited in claim 1, wherein the data units on any one of the storage devices are duplicated entirely on another single respective storage device.
Claim 6
Dependent Claim
79 words
A method as recited in claim 1, wherein the latency time is determined by measuring the angular position of a corresponding storage device at time t and using the following formula, which yields the angular position of the corresponding storage device at time T: [ n + (T - t) ] mod M where x is the rotation time, n is the angular position of the mechanism at time t, and M is the number of sectors per track.
Claim 7
Dependent Claim
111 words
A method as recited in claim 1, wherein each request to store data comprises a primary write request and a secondary write request, the primary request being a request to originally store data on a first storage device and the secondary write request being a request to store a copy of the originally stored data on a second storage device which is different than the first storage device, with the first and second storage devices being part of the plurality of storage devices, the secondary write request being placed in a low priority queue containing requests which are satisfied only if there are no requests in a respective higher priority queue.
Claim 8
Dependent Claim
190 words
In a direct access storage system with a plurality of circulating storage devices, each device having an associated high and low priority queue, a method of decreasing response time for storing or reading data, comprising the steps of: (a) estimating access time for satisfying each request of a plurality of requests, the requests being either to store or read data; (b) storing first requests (of said plurality of requests) to store data in respective high priority queues and second requests (of said plurality of requests) in respective low priority queues, each second request (of the second requests) being a request to store the same data as requested for storage by a corresponding first request (of the first requests), with each first request estimated to require less access time than a corresponding second request; (c) storing requests (of said plurality of requests) to access data, in respective queues of the associated high priority queues; and (d) servicing the requests stored in the low and high priority queues, wherein for each storage device, the requests in an associated high priority queue are serviced before the requests in an associated low priority queue.
Claim 9
Dependent Claim
190 words
In a direct access storage system with a plurality of circulating storage devices, each device having an associated high and low priority queue, a method of decreasing response time for storing or reading data, comprising the steps of: (a) estimating access time for satisfying each request of a plurality of requests, the requests being either to store or read data; (b) storing first requests (of said plurality of requests) to store data in respective high priority queues and second requests (of said plurality of requests) in respective low priority queues, each second request (of the second requests) being a request to store the same data as requested for storage by a corresponding first request (of the first requests), with each first request estimated to require less access time than a corresponding second request; (c) storing requests (of said plurality of requests) to access data, in respective queues of the associated high priority queues; and (d) servicing the requests stored in the low and high priority queues, wherein for each storage device, the requests in an associated high priority queue are serviced before the requests in an associated low priority queue.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.