US4858110A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Parallel data processor with shift register inter-processor transfer
Assignee: PRESIDENT OF AGENCY OF INDUSTRIAL SCIENCE & TECHNOLOGY
•
Filing: 1987-03-24
•
Granted: 1989-08-15
•
Expires: 2007-03-24
Abstract
Abstract A parallel data processor is disclosed, in which a plurality of fundamental computing elements each comprising a computer and a memory are connected in a two-dimensional lattice configuration, each of said fundamental computing elements being operated by an identical instruction provided by a single control section, and in which said each fundamental computing element is provided with a shift register having ability of data transmission to the memory in said element and of further data transmission, which is asynchronous to operation in said element, between the adjacent fundamental computing elements, said latter data transmission being conducted concurrently to the operation in said element.
Inventors
- 👤 Hiroyuki Miyata
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-27 12:05
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/76; G06F15/80; G06F15/8007; G06F15/8023
Claims Analysis
Total Claims: 5
Claim 1
Independent Claim
278 words
A parallel data processor comprising: a plurality of fundamental computing elements (1) interconnected in a two-dimensional lattice configuration, each computing element (1) further comprising: a computer unit (2) including a memory unit (3) and having a first speed of operation for processing data; a shift register (5) connected to said computer unit (2) for transmitting data therebetween and connected to shift registers of a plurality of adjacent fundamental computing elements (1) in the lattice configuration for transmitting data therebetween; a first clock means (6) for controlling said shift register (5) to transmit data between said shift register (5) and said memory unit (3) of computer unit (3) at a first rate that is synchronous to the first speed of operation of said computer unit (2), said first clock means (6) being connected to said computer unit (2) and said shift register (5); a second clock means (7) for controlling said shift register (5) to transmit data between said shift registers of said plurality of adjacent fundamental computing elements (1) at a second rate that is asynchronous to and faster than said first rate; a selector means (8) for enabling one of said first or second clock means (6,7) to control said shift register (5) so that data transmission within each fundamental computing element occurs at said first rate synchronously with said computer unit operation and so that data transmission between each said adjacent fundamental computing element occurs at said second rate asynchronously and independently of said computer unit operation; and a control station (20) for providing an identical instruction to each of said plurality of said fundamental computing elements (1) for operating said plurality of fundamental computing elements.
Claim 2
Dependent Claim
278 words
A parallel data processor comprising: a plurality of fundamental computing elements (1) interconnected in a two-dimensional lattice configuration, each computing element (1) further comprising: a computer unit (2) including a memory unit (3) and having a first speed of operation for processing data; a shift register (5) connected to said computer unit (2) for transmitting data therebetween and connected to shift registers of a plurality of adjacent fundamental computing elements (1) in the lattice configuration for transmitting data therebetween; a first clock means (6) for controlling said shift register (5) to transmit data between said shift register (5) and said memory unit (3) of computer unit (3) at a first rate that is synchronous to the first speed of operation of said computer unit (2), said first clock means (6) being connected to said computer unit (2) and said shift register (5); a second clock means (7) for controlling said shift register (5) to transmit data between said shift registers of said plurality of adjacent fundamental computing elements (1) at a second rate that is asynchronous to and faster than said first rate; a selector means (8) for enabling one of said first or second clock means (6,7) to control said shift register (5) so that data transmission within each fundamental computing element occurs at said first rate synchronously with said computer unit operation and so that data transmission between each said adjacent fundamental computing element occurs at said second rate asynchronously and independently of said computer unit operation; and a control station (20) for providing an identical instruction to each of said plurality of said fundamental computing elements (1) for operating said plurality of fundamental computing elements.
Claim 3
Independent Claim
241 words
A method for transmitting data in a parallel data processor having a plurality of fundamental computing elements (1) interconnected in a two-dimensional lattice configuration, each fundamental computing element (1) having a computer unit (2) including a memory unit (3) for storing data and having a first speed of operation for processing data, and a shift register (5) for transmitting data to and from said memory (3) at a first speed corresponding to the speed of operation of the computer unit (2) and to and from a second plurality of adjacent fundamental computing elements (1) in the lattice configuration at second speed, said second speed different from said first speed, the method comprising the steps of: operating the computer unit (2) of a first fundamental computing element (1) at said first speed; transmitting data between a memory (3) of said first fundamental computing element (1) and the shift register (5) of said first fundamental computing element (1) at said first speed; transmitting data between the shift register (5) of said first fundamental computing element and the shift registers of said second plurality of fundamental computing elements adjacent to said first fundamental computing element (1) at said second speed; and continuing operation of the computer unit (2) of the first fundamental computing element at said first speed independently of and concurrently with the transmission of data between the shift registers of said first and said second fundamental computing elements at said second speed.
Claim 4
Independent Claim
241 words
A method for transmitting data in a parallel data processor having a plurality of fundamental computing elements (1) interconnected in a two-dimensional lattice configuration, each fundamental computing element (1) having a computer unit (2) including a memory unit (3) for storing data and having a first speed of operation for processing data, and a shift register (5) for transmitting data to and from said memory (3) at a first speed corresponding to the speed of operation of the computer unit (2) and to and from a second plurality of adjacent fundamental computing elements (1) in the lattice configuration at second speed, said second speed different from said first speed, the method comprising the steps of: operating the computer unit (2) of a first fundamental computing element (1) at said first speed; transmitting data between a memory (3) of said first fundamental computing element (1) and the shift register (5) of said first fundamental computing element (1) at said first speed; transmitting data between the shift register (5) of said first fundamental computing element and the shift registers of said second plurality of fundamental computing elements adjacent to said first fundamental computing element (1) at said second speed; and continuing operation of the computer unit (2) of the first fundamental computing element at said first speed independently of and concurrently with the transmission of data between the shift registers of said first and said second fundamental computing elements at said second speed.
Claim 5
Dependent Claim
41 words
The method of claim 2 further comprising transmitting the data from the shift register of one of said second plurality of adjacent fundamental computing elements to a memory unit of said one adjacent second fundamental computing element at said first speed.
References Cited (Backward Citations)
US3582899A
Cited
US4174514A
Cited
US4215401A
Cited
US4380046A
Cited
JPS56123069A
Cited
US4418383A
Cited
US4580215A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1992-08-14 | 1993-02-13 | 1993-08-15 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1996-08-14 | 1997-02-13 | 1997-08-15 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2000-08-14 | 2001-02-13 | 2001-08-15 | $7700.00 | Window Closed / Lapsed |