Distributed data exchange system and distributed data exchange method based on RK3399Pro

Abstract

The invention provides a distributed data exchange system and a distributed data exchange method based on RK3399Pro. The invention comprises a data exchange front-end server connected to an external network server and an isolation gatekeeper, the data exchange front-end server comprises a processing unit. The processing unit comprises a distributed processing module, a distributed slot module anda switch module; the distributed processing module comprises a plurality of core modules. Each core module comprises an AI processor RK3399Pro and a peripheral circuit connected to the AI processor RK3399Pro. The distributed slot module comprises a plurality of slot units. Each core module is connected to one slot unit, the switch module is connected to the isolation gatekeeper and the extranet server, the switch module comprises a switch mainboard and a network controller connected to the switch mainboard, the network controller is used for data exchange scheduling and data acquisition management, and the switch mainboard is connected to the plurality of slot units. The invention has the advantages of excellent performance, stable work, safe data, easy expansion, high cost performance andthe like.

Description

technical field [0001] The invention belongs to the technical field of data exchange equipment, and in particular relates to a distributed data exchange system and exchange method based on RK3399Pro. Background technique [0002] In recent years, with the acceleration of network security threats, it is necessary to consider solving core data security issues in the construction of network information security systems. Now enterprises and institutions will consider separating internal and external networks in their network construction. Internet and intranet safety measures such as isolation. This has also prompted the demand for data exchange equipment on the internal and external networks to enter a stage of rapid development. With the advancement of informatization construction and the gradual deepening of the digital transformation of enterprises and institutions, it is increasingly necessary to carry out continuous large-scale data exchange between internal and external ...

AI Technical Summary

Problems solved by technology

[0005] When the network transmission traffic is high, especially when the traffic reaches several gigabytes or even 10 gigabytes, and there are a large number of concurrent business flows, the amount of data to be processed per unit time is more, and the TCP / IP protocol stack consumes high CPU resources. The characteristics of the server system will cause a very high CPU load, which will greatly reduce the efficiency of the server to read and write data.

A large number of soft interrupts generated by high concurrency will also reduce system processing speed and affect work efficiency

At present, in order to improve the efficiency of data exchange, the common method is to improve the performance of system hardware and adopt higher-performance multi-core servers and network equipment, but the cost is to modify the system to a certain extent. For some large systems, the cost is very high. huge

[0006] 2. Data security issues

[0007] The X86 architecture has been developed for a long time and has a high market share, so devices based on this architecture are well known in the industry, including systems, interfaces, storage, etc. Therefore, data is relatively more likely to be infected by viruses, illegally copied, etc.

[0008] 3. Poor scalability

[0009] Generally, the upgrade of equipment can only be upgraded in terms of the number or performance of memory and hard disks. CPUs can only be upgraded in the same series of products, and the number of processor systems cannot be easily upgraded. The overall performance scalability is limited.

[0010] 4. The cost and power consumption are relatively large

[0011] Due to the large amount of data traffic accessed by internal and external networks in the application, and the amount of parallel access to establish links is also large, the requirements for the main frequency of the CPU and the number of threads are relatively high, which requires a high-performance server CPU, such as: Intel's Xeon series processors, but its price is relatively high, several times that of ARM processors

In order to improve the processing capacity of data exchange, when using dual CPUs or adding equipment, the price of the product will increase exponentially

At the same time, the operation of Intel's processor also requires additional memory, hard disk, graphics card, CPU cooling fan, high-power power supply, etc. The cost of these devices is also very high.

[0012] In addition, the X86 architecture CPU emphasizes the idea of ​​performance + speed, which also leads to a relatively large power consumption of the CPU. The power consumption of an internal and external network data exchange device with a CPU can reach several hundred watts, and the volume of the CPU and the device is relatively large.

Images