Distributed system, main system and slave systems thereof and a code loading method

Abstract

The invention discloses a distributed system, a main system and slave systems thereof and a code loading method. In the distributed system, the main system is used for uniformly managing and maintaining a guidance procedure and an application procedure of each slave system, and a double-interface RAM (random access memory) used for storing the guidance procedure of each slave system is formed in a logic chip of each slave system, so that the main system can respectively write the guidance procedure of each slave system into the double-interface RAM of the logic chip of the slave system and also can carry out reset removal to a CPU (Central Processing Unit) of the slave system after the completion of writing to lead the CPU of the slave system to load the guidance procedure of the system from the double-interface RAM of the logic chip of the slave system, thereby uniformly realizing the loading of the guidance procedure of each slave system and the initial start-up through the main system, reducing the production and maintenance cost of the systems, improving the reliability and the robustness of the systems, reducing the limitation during the upgrade and expansion process, and reducing the energy consumption of the systems.

Description

technical field [0001] The present invention relates to code loading technology, in particular to a distributed system that can realize code loading of each slave device, a master system that can load code to a slave system in a distributed system, and a master system that can load code from a distributed system A slave system, and a code loading method in a distributed system. Background technique [0002] A distributed system generally consists of a master device and at least one slave system. Each slave system and the master system respectively execute specific applications and communicate through inter-system interfaces to complete specific functions in cooperation. Wherein, both the master system and the slave system include CPUs such as PCI controllers, random memory devices such as SDRAM, programmable logic chips such as FPGA / CPLD, and non-volatile memory devices. [0003] During the start-up process of the distributed system, each CPU of the slave system first reads...

AI Technical Summary

Problems solved by technology

[0006] 1. Each slave system needs at least one non-volatile storage device, and at least needs to pre-write the boot program of the system in the non-volatile storage device, which increases the production cost of the slave system;

[0007] 2. Non-volatile storage devices and their wiring will take up a lot of space, which is not conducive to the miniaturization of the system;

[0008] 3. There is also a version matching relationship between the boot program and the application program of each slave system, thus increasing the maintenance cost of the slave system;

[0009] 4. In order to ensure the version matching relationship, it is easy to make mistakes when manually upgrading the boot program of the slave system. For example, a power failure occurs when the boot program is being written to the non-volatile storage device of the slave system, which may cause the slave system to fail to start normally;

[0010] 5. Upgrading and business expansion from the subsequent version of the system boot program and application program are limited by the capacity of the non-volatile storage device;

[0011] 6. The non-volatile storage device of the slave system is still powered on to consume energy after startup, which increases the total power consumption of the distributed system

[0012] It can be seen that the existing distributed system has many defects such as high cost, low reliability, limited upgrade and expansion, and high power consumption.

Images