Heterogeneous multi-core parallel processing device and method for heterogeneous multi-source big data

Abstract

The invention discloses a heterogeneous multi-core parallel processing device and method for heterogeneous multi-source big data. The processing device and method are realized by the adoption of a ZYNQ-7000SoC chip of the Xilinx company. The device comprises an FPGA-based heterogeneous multi-source big data parallel collection module and a heterogeneous multi-core data parallel processing module,wherein the parallel collection module caches collected data of different equipment into an off-chip DDR with different offset addresses and sizes through an on-chip HP port in a DMA mode, and meanwhile a design state machine mounts the off-chip DDR on an on-chip AXI bus; and multiple MicroBlaze cores are set up in the parallel processing module, and the MicroBlaze cores and an ARM core form a heterogeneous multi-core architecture, wherein different MicroBlaze cores are responsible for processing different equipment data, the ARM core completes performance monitoring on the MicroBlaze cores, and different data processing algorithms are dynamically scheduled to run on the MicroBlaze cores to guarantee load balance of the cores. Through the processing device and method, efficient parallel processing of big data of different equipment on a manufacturing site can be realized, and an intelligent manufacturing superior decision can be effectively supported.

Description

technical field [0001] The invention belongs to the fields of electronic engineering and computer science, and in particular relates to a heterogeneous multi-core parallel processing device and method for heterogeneous multi-source big data. Background technique [0002] With the proposal of the national strategy "Made in China 2025", intelligent manufacturing has become a hot term in contemporary China. However, the realization of intelligent manufacturing is inseparable from data. These data are more specifically represented as the data of the manufacturing site, and a wide variety of equipment constitutes the manufacturing site, such as supporting industrial Ethernet interfaces, RS232 interfaces, asynchronous RS422 interfaces, asynchronous RS485 interfaces, and synchronous RS485 interfaces. Interface, SPI interface, I2C interface, CAN interface, MTConnect interface and OPC UA interface. In order to realize the effective monitoring of the manufacturing site, it is particu...

AI Technical Summary

Problems solved by technology

In order to realize the effective monitoring of the manufacturing site, it is particularly necessary to realize the parallel data collection and processing of various equipment on the site, while the traditional methods mostly use multiple processors / embedded microprocessors to separately collect and process different equipment data , and then these processors / embedded microprocessors perform data interaction through the off-chip bus, and this traditional method will inevitably increase the number of on-site acquisition and processing devices, increase the cost, and increase the difficulty of on-site layout and wiring At the same time, the data interaction based on the off-chip bus also increases the delay and reduces the real-time performance of data transmission and processing

Images