US5574933A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Task flow computer architecture
Assignee: Hewlett Packard Development Co LP
•
Filing: 1994-10-26
•
Granted: 1996-11-12
•
Expires: 2014-10-26
Abstract
Abstract A computer architecture has a plurality of processing cells interconnected to perform programming tasks. Each cell contains both memory and processing elements. Memory packets contain an instruction, a data element, and a pointer to another memory packet. Tasks are executed by following a linked list of memory packets. Transmission packets communicate instructions and register values along the linked list. A plurality of computer processes may be executed simultaneously.
Inventors
- 👤 Robert W. Horst
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 15:14
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/76; G06F15/80; G06F15/8007; G06F15/8015; G06F9/00; G06F9/06; G06F9/44; G06F9/448; G06F9/4494
Claims Analysis
Total Claims: 11
Claim 1
Independent Claim
365 words
A computer for executing a task, the task being segmented into a plurality of instructions to be executed sequentially, said computer comprising: (a) an interconnection network; (b) a first cell coupled to said interconnection network, said first cell including: a first receive queue coupled to said interconnection network for receiving a first incoming transmission packet containing a first instruction and a first link address, said first link address identifying a first memory packet and forming a portion of a linked list of memory packets specifying an order of execution; a first memory coupled to said first receive queue for storing said first memory packet, said first memory packet being selectively disposed in said first cell and including a second instruction, a first data element and a second link address, said second link address identifying a second memory packet and forming a portion of said linked list of memory packets specifying an order of execution; a bypass queue coupled to said first memory for outputting a second incoming transmission packet that addresses no memory packet residing in said first memory; a first arithmetic and logic unit (ALU) coupled to said first memory for executing said first instruction; means coupled to said first ALU for generating a first outgoing transmission packet containing said second instruction and a link field for storing said second link address; and (c) a second cell coupled to said first cell through said interconnection network, said second cell comprising: a second receive queue coupled to said interconnection network for receiving said first outgoing transmission packet; a second memory coupled to said second receive queue for storing said second memory packet, said second memory packet being selectively disposed in said second cell and including a third instruction, a second data element and a third link address, said third link address identifying a third memory packet and forming a portion of said linked list of memory packets specifying an order of execution; a second ALU coupled to said second memory for executing said second instruction stored in said first outgoing transmission packet; and means coupled to said second ALU for generating a second outgoing transmission packet containing said third instruction and said third link address.
Claim 2
Dependent Claim
365 words
A computer for executing a task, the task being segmented into a plurality of instructions to be executed sequentially, said computer comprising: (a) an interconnection network; (b) a first cell coupled to said interconnection network, said first cell including: a first receive queue coupled to said interconnection network for receiving a first incoming transmission packet containing a first instruction and a first link address, said first link address identifying a first memory packet and forming a portion of a linked list of memory packets specifying an order of execution; a first memory coupled to said first receive queue for storing said first memory packet, said first memory packet being selectively disposed in said first cell and including a second instruction, a first data element and a second link address, said second link address identifying a second memory packet and forming a portion of said linked list of memory packets specifying an order of execution; a bypass queue coupled to said first memory for outputting a second incoming transmission packet that addresses no memory packet residing in said first memory; a first arithmetic and logic unit (ALU) coupled to said first memory for executing said first instruction; means coupled to said first ALU for generating a first outgoing transmission packet containing said second instruction and a link field for storing said second link address; and (c) a second cell coupled to said first cell through said interconnection network, said second cell comprising: a second receive queue coupled to said interconnection network for receiving said first outgoing transmission packet; a second memory coupled to said second receive queue for storing said second memory packet, said second memory packet being selectively disposed in said second cell and including a third instruction, a second data element and a third link address, said third link address identifying a third memory packet and forming a portion of said linked list of memory packets specifying an order of execution; a second ALU coupled to said second memory for executing said second instruction stored in said first outgoing transmission packet; and means coupled to said second ALU for generating a second outgoing transmission packet containing said third instruction and said third link address.
Claim 3
Dependent Claim
15 words
The computer of claim 1 wherein said first input queue comprises a bidirectional shift register.
Claim 4
Dependent Claim
14 words
The computer of claim 1 wherein said interconnection network comprises a two-bus bidirectional ring.
Claim 5
Dependent Claim
14 words
The computer of claim 1 wherein said interconnection network comprises a four-bus bidirectional ring.
Claim 6
Dependent Claim
13 words
The computer of claim 1 wherein said interconnection network comprises a unidirectional Torus.
Claim 7
Dependent Claim
20 words
The computer of claim 1 further comprising a global synchronous clock coupled to said first cell and said second cell.
Claim 8
Dependent Claim
42 words
The computer of claim 6 further comprising a watchdog timer in said first cell, said timer generating a signal indicating that said first incoming receive queue has not changed state within a predetermined number of clock cycles of said global synchronous clock.
Claim 9
Dependent Claim
33 words
The computer of claim 1 wherein said interconnection network has a first width and said first incoming outgoing transmission packet has a second width, said first width being different from said second width.
Claim 10
Dependent Claim
15 words
The computer of claim 8 wherein said first width is less than said second width.
Claim 11
Dependent Claim
25 words
The computer of claim 9 wherein said interconnection network forwards a first incoming portion and a second portion of said first outgoing transmission packet serially.
References Cited (Backward Citations)
US3962706A
Cited
US4149240A
Cited
US4153932A
Cited
US4837676A
Cited
US5133062A
Cited
US5021947A
Cited
US4893234A
Cited
US4972315A
Cited
US5117489A
Cited
US5093919A
Cited
US5043880A
Cited
US5179685A
Cited
US5095480A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-11-12 | 2000-05-13 | 2000-11-12 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-11-12 | 2004-05-13 | 2004-11-12 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-11-12 | 2008-05-13 | 2008-11-12 | $7700.00 | Window Closed / Lapsed |