Patent US2810950A successfully refreshed from Google Patents.
Patent US4514807A successfully refreshed from Google Patents.
US4514807A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Parallel computer
Assignee: Individual
•
Filing: 1984-02-13
•
Granted: 1985-04-30
•
Expires: 2004-02-13
Abstract
Abstract A parallel computer of a MIMD array type suitable for high speed computations in engineering and science, comprising a master unit, N submaster units and a two dimensional array of N 2 arithmetic units having own processors and private memories respectively, and further N common memories each of which is used for transferring some data between a submaster and a corresponding set of N arithmetic units, and finally a three dimensional array of buffer memories with a capacity being composed of N grand boards each (alternatively the k-th grand board) of which has N 2 memory units and is provided for transferring some data between every pair of an arithmetic unit being on a (the k-th) row of the arithmetic unit array and another unit being on the corresponding (the k-th) column.
Inventors
- 👤 Tatsuo Nogi
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-03 19:21
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/16; G06F15/163; G06F15/173; G06F15/17356; G06F15/17368; G06F15/17393; G06F15/76; G06F15/80; G06F15/8007; G06F15/8023
Claims Analysis
Total Claims: 3
Claim 1
Independent Claim
391 words
A multiprocessor parallel computer system comprising: a control computer system; a plurality of identical slave processors arranged in an N by N matrix in the form of ______________________________________
AU-((0,0)), AU-((0,1)), . . . , AU-((0,N-1))
AU-((1,0)), Au-((1,1)), . . . , AU-((1,N-1))
. . . AU-((j,k)) . . .
. . . AU-((k,i)) . . .
AU-((N-1,0)), AU-((N-1,1)), . . . , AU-((N-1,N-1))
______________________________________ where N is an integer selected to be greater than one, and where AU-((j,k)) is the slave processor located in the jth row and kth column of said N by N matrix, each row of N of said slave processors being connected to said control computer system through a separate bus serving said row of N slave processors; and a plurality of identical buffer memory units arranged in a three-dimensional array consisting of N matricies, N by N, in the form of ______________________________________
(0,0).sub.k ; (0,1).sub.k ; . . . ; (0,N-1).sub.k
(1,0).sub.k ; (1,1).sub.k ; . . . ; (1,N-1).sub.k
. . . (i,j).sub.k . . .
. . . (j,i).sub.k . . .
(N-1,0).sub.k ; (N-1,1).sub.k ; . . . ; (N-1,N-1).sub.k
______________________________________ where the index k=0, 1, 2, . . . , N-1 is an array number, and where any buffer memory unit (i,j) k located in the jth column and the ith row of the kth matrix of said N by N memory matricies, and every buffer memory unit on the ith row, (i,j) k (j=0,1, . . . , N-1) is interconnected to the processor AU-((k,i)) via a corresponding common `row` bus, while every unit on the jth column, (i,j) k (i=0,1,2, . . . , N-1) is interconnected to the processor AU-((j,k)) via a corresponding common `column` bus; and where any two processors with the same index k, said AU-((k,i)) and said AU((j,k)), can transfer data to one another through the buffer memory unit (i,j) k , and any two slave processors, AU-((l,m)) and said AU-((j,k)), not directly interconnected by a corresponding buffer memory unit, transfer data to one another through either of two mediation slave processors, AU-((m,j)) or AU-((k,l)), said slave processors AU-((m,j)) and AU-((k,l)) being interconnected to said slave processor AU-((l,m)) through corresponding buffer memory units (j,l)m and (m,k) l , respectively, and to said processor AU-((j,k)), through corresponding buffer memory units, (k,m) j and (l,j) k , respectively.
Claim 2
Dependent Claim
391 words
A multiprocessor parallel computer system comprising: a control computer system; a plurality of identical slave processors arranged in an N by N matrix in the form of ______________________________________
AU-((0,0)), AU-((0,1)), . . . , AU-((0,N-1))
AU-((1,0)), Au-((1,1)), . . . , AU-((1,N-1))
. . . AU-((j,k)) . . .
. . . AU-((k,i)) . . .
AU-((N-1,0)), AU-((N-1,1)), . . . , AU-((N-1,N-1))
______________________________________ where N is an integer selected to be greater than one, and where AU-((j,k)) is the slave processor located in the jth row and kth column of said N by N matrix, each row of N of said slave processors being connected to said control computer system through a separate bus serving said row of N slave processors; and a plurality of identical buffer memory units arranged in a three-dimensional array consisting of N matricies, N by N, in the form of ______________________________________
(0,0).sub.k ; (0,1).sub.k ; . . . ; (0,N-1).sub.k
(1,0).sub.k ; (1,1).sub.k ; . . . ; (1,N-1).sub.k
. . . (i,j).sub.k . . .
. . . (j,i).sub.k . . .
(N-1,0).sub.k ; (N-1,1).sub.k ; . . . ; (N-1,N-1).sub.k
______________________________________ where the index k=0, 1, 2, . . . , N-1 is an array number, and where any buffer memory unit (i,j) k located in the jth column and the ith row of the kth matrix of said N by N memory matricies, and every buffer memory unit on the ith row, (i,j) k (j=0,1, . . . , N-1) is interconnected to the processor AU-((k,i)) via a corresponding common `row` bus, while every unit on the jth column, (i,j) k (i=0,1,2, . . . , N-1) is interconnected to the processor AU-((j,k)) via a corresponding common `column` bus; and where any two processors with the same index k, said AU-((k,i)) and said AU((j,k)), can transfer data to one another through the buffer memory unit (i,j) k , and any two slave processors, AU-((l,m)) and said AU-((j,k)), not directly interconnected by a corresponding buffer memory unit, transfer data to one another through either of two mediation slave processors, AU-((m,j)) or AU-((k,l)), said slave processors AU-((m,j)) and AU-((k,l)) being interconnected to said slave processor AU-((l,m)) through corresponding buffer memory units (j,l)m and (m,k) l , respectively, and to said processor AU-((j,k)), through corresponding buffer memory units, (k,m) j and (l,j) k , respectively.
Claim 3
Dependent Claim
50 words
A computer system as recited in claim 1 wherein said control computer system comprises a master processor and a plurality, N, of submaster processors, each of said submaster processors being connected to said master processor and to a corresponding row of said slave processors of said matrix of slave processors.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1988-04-29 | 1988-10-29 | 1989-04-30 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1992-04-29 | 1992-10-29 | 1993-04-30 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1996-04-29 | 1996-10-29 | 1997-04-30 | $7700.00 | Window Closed / Lapsed |