Convex Computer Corporation

Other

James E Mankovich

Gary Belgrave Gostin

David M Chastain

Steven Jeffrey Wallach

Thomas M Jones

Harold Wallace Dozier

Arthur T Kimmel

John W Clark

James R Weatherford

Alan D Gant

David A Nobles

Jeffrey H Gruger

Michael C Harris

Kent A Fuka

Frank J Marshall

Harold D Shelton

Gilbert J Hansen

John M Golenbieski

Gaynel J Lockhart

Gerald Matulka

Michael E Sydow

Larry W Spry

Robert B Kolstad

William H Wallace

Steven M Rowan

A computer system having a plurality of independent processors which can either execute a separate process for each processor, or execute parallel process operations across multiple processors for one process. The computer system includes a set of communication registers divided into a group of frames and a set of semaphores which correspond respectively to the registers. Typical processes have both serial and parallel code segments. During serial execution, a process is executed by a single processor, but when a parallelization instruction is encountered, which indicates that code can be executed in parallel, a semaphore is posted to invite other processors to join in parallel execution of the process. If any other processors in the system are idle, those processors detect the semaphore and accept a thread of process operation. Two or more processors may join in parallel operation if sufficient operations are available. However, if all processors are busy, then the processor conducting the serial operation will also execute all of the parallel operations. Thus, the processors are self-allocated, rather than being used on a demand or master-servant basis. This permits a greater throughput for the system because processors are not held idle when there is available process work to be done.

Multi-processor computer system having process-independent communication

A computer system comprises a plurality of independent processors which can either execute a separate process for each processor, or execute parallel process operations across multiple processors for one process. The computer system includes a set of communication registers divided into a group of frames and a set of semaphores which correspond respectively to the registers. Typical processes have both serial and parallel code segments. During serial execution, a process is executed by a single processor, but when a parallelization instruction is encountered, which indicates that code can be executed in parallel, a semaphore is posted to invite other processors to join in parallel execution of the process. If any other processors in the system are idle, those processors detect the semaphore and accept a thread of process operation. Two or more processors may join in parallel operation if sufficient operations are available. However, if all processors are busy, then the processor conducting the serial operation will also execute all of the parallel operations. Thus, the processors are self-allocated, rather than being used on a demand or master-servant basis. This permits a greater throughput for the system because processors are not held idle when there is available process work to be done.

Multi-processor computer system having self-allocating processors

A vector processing computer is configured to operate in a pipelined fashion wherein each of the functional units is essentially independent and is designed to carry out its operational function in the fastest possible manner. Vector elements are transmitted from memory, either main memory, a physical cache unit or a logical cache through a source bus where the elements are alternately loaded into the vector processing units. The vector control unit decodes the vector instructions and generates the required control commands for operating the registers and logical units within the vector processing units. Thus, the vector processing units essentially work in parallel to double the processing rate. The resulting vectors are transmitted through a destination bus to either the physical cache unit, the main memory, the logical cache or to an input/output processor. In a further aspect of the computer there is produced an entry microword from a store for the immediate execution of the first microinstruction within a sequence of microinstructions. The remaining microinstructions are produced from a conventional store. This reduces the delay in the retrieval and execution of the first microinstruction. In a still further aspect of the computer there is included the logical data cache which stores data at logical addresses such that the central processor can store and retrieve data without the necessity of first making a translation from logical to physical address.

Computer system

Growing community of inventors

Colorado Springs, CO, United States of America

James E Mankovich