Big data processing device

Abstract

The invention provides a big data processing device. The device consists of a power supply, an interface unit, a preprocessing unit, a process scheduling chip and a CPU unit; the power supply is a dual-redundancy power supply and supplies power to the interface unit, the preprocessing unit, the process scheduling chip and the CPU; the preprocessing unit comprises multiple FPGA chips; and the CPU unit comprises multiple embedded CPUs. The multiple FPGA chips are added for performing preprocessing; the multiple CPUs are flexibly configured for performing multi-kernel processing according to dataflow; data collection, data filtering and data processing are integrated in one device, so that the processing efficiency is greatly improved and the processing speed is increased; and the device isespecially suitable for the industries with high data processing delay requirements, such as finance, securities and the like.

Description

technical field [0001] The invention relates to a big data processing device. Background technique [0002] Commonly used data processing equipment adopts the way of integrating network cards in servers. The servers are bulky and power-hungry; the network cards are limited by the size of the server’s chassis and the traditional CPU I / O bus. There are few data ports and low processing efficiency. [0003] In high-speed data processing, distributed deployment is often used. When necessary, data distribution equipment is added to the front end to preprocess the data. The system is huge, which not only fails to meet the requirements of energy saving and emission reduction, but also increases the cost of the system. Contents of the invention [0004] The purpose of the present invention is to solve the problems existing in the prior art, and provide a large data processing device with good processing effect, small size, light weight, low power consumption, and portability. [...

AI Technical Summary

Problems solved by technology

[0002] The commonly used data processing equipment adopts the way of integrating the network card in the server. The server is bulky and consumes power; the network card is limited by the size of the server’s chassis and the traditional CPU I / O bus. There are few data ports and low processing efficiency.

[0003] In high-speed data processing, distributed deployment is often used. When necessary, data distribution equipment is added to the front end to preprocess the data. The huge system not only fails to meet the requirements of energy saving and emission reduction, but also increases the cost of the system.

Images