Patent US4412303A successfully refreshed from Google Patents.
US4412303A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Array processor architecture
Assignee: Unisys Corp
•
Filing: 1979-11-26
•
Granted: 1983-10-25
•
Expires: 1999-11-26
Abstract
Abstract A high speed parallel array data processing architecture fashioned under a computational envelope approach includes a data base memory for secondary storage of programs and data, and a plurality of memory modules interconnected to a plurality of processing modules by a connection network of the Omega gender. Programs and data are fed from the data base memory to the plurality of memory modules and from hence the programs are fed through the connection network to the array of processors (one copy of each program for each processor). Execution of the programs occur with the processors operating normally quite independently of each other in a multiprocessing fashion. For data dependent operations and other suitable operations, all processors are instructed to finish one given task or program branch before all are instructed to proceed in parallel processing fashion on the next instruction. Even when functioning in the parallel processing mode however, the processors are not locked-step but execute their own copy of the program individually unless or until another overall processor array synchronization instruction is issued.
Inventors
- 👤 George H. Barnes
- 👤 Stephen F. Lundstrom
- 👤 Philip E. Shafer
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-27 23:50
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
Claims Analysis
Total Claims: 9
Claim 1
Independent Claim
319 words
An array processing system interfacing with a host processor for performing I/O functions and a file memory, said array processing system comprising: data base buffer memory means having at least one data transfer channel connected to the file memory under control of the support processor for rapid transfer of information therethrough, said data base memory means receiving parallel processing jobs from the file memory and returning to the file memory the processed resultants of the jobs received, said parallel processing jobs including both programs for vector and scalar processing and related data; an array of parallel memory modules connected via a second data transfer channel to said data base memory means, each memory module in said array storing data from said data base memory means during processing by said array processing system; an array of parallel processor means, each processor means being programmable for independently processing data fetched approximately in parallel from individual memory modules, each processor means having indicator means for indicating when a future operation must be initiated in parallel with at least one other processor means for indicating when it is ready to begin said future operation; connection network means interposing said array of parallel memory modules and said array of parallel processor means, said connection network means providing a plurality of data communication paths between said array of parallel memory modules and said array of parallel processor means wherein each provided data communications path in said plurality thereof is provided to a requested individual memory module from a requesting individual processor means; and coordinator means connected to each processor means for monitoring each said indicator means for indicating when a future operation must be initiated in parallel with at least one other processor means and for inhibiting each processor means so indicating from initiating said future operation until each processor means so indicating is also indicating that it is ready to begin it's future operation.
Claim 2
Dependent Claim
319 words
An array processing system interfacing with a host processor for performing I/O functions and a file memory, said array processing system comprising: data base buffer memory means having at least one data transfer channel connected to the file memory under control of the support processor for rapid transfer of information therethrough, said data base memory means receiving parallel processing jobs from the file memory and returning to the file memory the processed resultants of the jobs received, said parallel processing jobs including both programs for vector and scalar processing and related data; an array of parallel memory modules connected via a second data transfer channel to said data base memory means, each memory module in said array storing data from said data base memory means during processing by said array processing system; an array of parallel processor means, each processor means being programmable for independently processing data fetched approximately in parallel from individual memory modules, each processor means having indicator means for indicating when a future operation must be initiated in parallel with at least one other processor means for indicating when it is ready to begin said future operation; connection network means interposing said array of parallel memory modules and said array of parallel processor means, said connection network means providing a plurality of data communication paths between said array of parallel memory modules and said array of parallel processor means wherein each provided data communications path in said plurality thereof is provided to a requested individual memory module from a requesting individual processor means; and coordinator means connected to each processor means for monitoring each said indicator means for indicating when a future operation must be initiated in parallel with at least one other processor means and for inhibiting each processor means so indicating from initiating said future operation until each processor means so indicating is also indicating that it is ready to begin it's future operation.
Claim 3
Dependent Claim
58 words
The array processing system according to claim 1 wherein: said array of parallel memory modules comprises an array of a prime number of identical memory modules; and said array of parallel processor means comprises an array of a power-of-two number of identical processors, said power-of-two number being the greatest power-of-two number less in value than said prime number.
Claim 4
Dependent Claim
20 words
The array processing system according to claim 2 wherein said prime number is 521 and said power-of-two number is 512.
Claim 5
Dependent Claim
56 words
The array processing system according to claim 1 wherein said data base memory means includes: a solid-state information storage unit; and a data base memory controller means for providing buffering and interfacing between said solid-state information storage unit and the file memory and between said solid-state information storage unit and said array of parallel memory modules.
Claim 6
Dependent Claim
51 words
The array processing system according to claim 1 wherein each processor in said array of parallel processor means further includes: an execution unit; a processor memory for storing instructions for execution by said execution unit; and a connection network buffer for interfacing between said connection network means and said execution unit.
Claim 7
Dependent Claim
34 words
The array processing system according to claim 1 wherein each memory module in said array thereof includes a solid-state random access memory unit sufficient in size to store 64,000 words of 55 bits each.
Claim 8
Dependent Claim
40 words
The array processing system according to claim 1 wherein: said connection network further includes means for providing simultaneously data communication paths to all processor means in said array thereof for broadcasting thereto from any memory module in said array thereof.
Claim 9
Dependent Claim
43 words
The array processing system according to claim 1 or claim 7 wherein: said connection network further includes means for providing simultaneously data communication paths to all memory modules in said array thereof for broadcasting thereto from any processor means in said array thereof.
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) | 1986-10-24 | 1987-04-25 | 1987-10-25 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 1990-10-24 | 1991-04-25 | 1991-10-25 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 1994-10-24 | 1995-04-25 | 1995-10-25 | $7700.00 | Window Closed / Lapsed |