FIELD: computer technology.
SUBSTANCE: invention relates to the field of digital computing. The expected result is achieved by paralleling calculations to all available computing resources, such as processor cores of various architectures. According to the method, the parallel computing process of the central core and the computing process of the processor core are started. Messaging and synchronization of processes running on cores of different systems on a chip are carried out using the mechanism of local network sockets, and processes running on cores of the same system on a chip are carried out directly between computing processes using shared memory. Within each system, a common cyclic queue of local and global messages is organized on a chip. The message counter monitors the message lifecycle.
EFFECT: increase in the speed of data processing.
7 cl, 3 dwg
Title | Year | Author | Number |
---|---|---|---|
HIGH-PERFORMANCE COMPUTER PLATFORM BASED ON PROCESSORS WITH HETEROGENEOUS ARCHITECTURE | 2016 |
|
RU2635896C1 |
SMALL-SIZED HIGH-PERFORMANCE COMPUTING MODULE BASED ON A MULTIPROCESSOR SYSTEM-ON-A-CHIP | 2021 |
|
RU2778213C1 |
PROGRAM-HARDWARE PLATFORM AND METHOD OF ITS REALISATION FOR WIRELESS COMMUNICATION FACILITIES | 2016 |
|
RU2626550C1 |
COMPUTING DEVICE OF SOFTWARE AND HARDWARE COMPLEX | 2016 |
|
RU2618367C1 |
LAUNCHING APPLICATIONS ON BASIS OF MESSAGE TRANSMISSION INTERFACE (MPI) IN HETEROGENEOUS MEDIUM | 2014 |
|
RU2600538C2 |
COMPUTATIONAL MODULE FOR MULTI-STREAM PROCESSING OF DIGITAL DATA AND PROCESSING METHOD USING SAID MODULE | 2018 |
|
RU2708794C2 |
COMPUTER SYSTEM FOR SCIENTIFIC AND TECHNICAL CALCULATIONS | 2018 |
|
RU2710890C1 |
UNLIMITED TRANSACTIONAL MEMORY WITH ASSURANCES OF MOVEMENT DURING TRANSFER, USING HARDWARE GLOBAL LOCK | 2014 |
|
RU2597506C2 |
HIGH-PERFORMANCE COMPUTER SYSTEM AND METHOD | 2017 |
|
RU2733058C1 |
IMPROVEMENT OF PCI EXPRESS INTERFACE | 2013 |
|
RU2645288C2 |
Authors
Dates
2023-02-14—Published
2022-05-19—Filed