Programmable processing elements interconnected by a communication network including field operation unit for performing field operations

• US 5,093,920 A
• Filed: 09/15/1989
• Issued: 03/03/1992
• Est. Priority Date: 06/25/1987
• Status: Expired due to Fees

• Current Assignee: AT&T, Inc.
• Original Assignee: AT&T, Inc.
• Inventors: Dally, William J., Krishnakumar, Anjur S., Agrawal, Prathima
• Primary Examiner(s): Lee, Thomas C.
• Assistant Examiner(s): MOHAMED, AYNI Y
• Application Number:
  US07/408,459
• Time in Patent Office: 900 Days
• Field of Search: 364/200 MS File, 364/900 MS File
• US Class Current: 712/29
• CPC Class Codes: G06F 30/33 Design verification, e.g. functional simulation or model checking

Abstract

An arrangement for implementing graph operations, generally, and circuit simulations in particular employing a plurality of substantially identical high speed special purpose processing elements (PE) that are flexibly interconnected through a switch to form a cluster. In addition to the processing element (PE) being programmable, the interconnection switch permits dynamically altered routing of signals between the processing elements. The processing elements include a queue unit which permits high speed asynchronous communication between the elements. Additional advantage is attained with a hierarchical arrangement where a plurality of clusters are interconnected in an n-cube arrangement, and all of the clusters communicate with a host computer.

Claims

1. A programmable logic simulation accelerator comprising:

a plurality of programmable processing elements; and

a communication network for transferring information to and from said accelerator and between any one of said plurality of processing elements

with each of said plurality of processing elements including

a plurality of data buses

an input and output block for communicating messages between said plurality of data buses and said communication network, and

a field operation unit responsive to signals on said plurality of data buses for selecting a first chosen field from one of said plurality of data buses and a second chosen field from another one of said plurality of data buses, operating on the first and second chosen fields and placing a result in a preselected field on one of said plurality of data buses.

2. A processing element in accordance with claim 1, further including stall means for suspending selected operations when said queue is empty or full.

3. A processing element in accordance with claim 1, further including a memory and means for configuring said memory to any desired word width.

4. The accelerator of claim 1, further comprising:

an associated memory connected to each of said processing elements memory;

a common bus for interconnecting said memories; and

a common memory connected to said common bus

5. The accelerator of claim 4, further comprising a housekeeping processor interposed between said common bus and said communications network.

6. The accelerator of claim 1 further including means for designating said chosen field selected by said field operation unit from said one bus and said field chosen by said field operation unit from said another bus.