Multi-ring topology redundancy and data strong real-time fault tolerance method of optical fiber reflection memory network

Abstract

The invention discloses a multi-ring topology redundancy and data strong real-time fault-tolerant method for an optical fiber reflection memory network, which comprises the following steps of: inserting a reflection memory card into each computer, and connecting the reflection memory cards through an optical fiber to form a ring network; on the basis of a ring network topology structure, a dual-redundancy system with one main redundancy and one standby redundancy is adopted; the design subsystem is connected with a switch through a bridge node to realize cross-region strong real-time and quasi-synchronization; designing main and standby systems to transmit and receive the same data at the same time, and performing redundancy elimination on the data by a receiving end; a monitoring mechanism of the backup system is designed, and whether data transmission interruption occurs in the main system is judged; whether a fault exists or not is judged by detecting a data flow state and an optical signal of each port, and a redundant backup system is selected for fault tolerance; and finally, a standard clock source is adopted to share standard time to the whole network, and the nodes carry out time correction by reading the standard time. The invention provides high reliability and real-time performance, and ensures that the system can still keep data integrity under the condition of a fault.

Description

Technical Field

[0001] The present invention belongs to the technical field of reflective memory networks, and in particular relates to a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network. Background Art

[0002] Fiber Optic Reflective Memory (FORM) is a high-speed, high-reliability, and low-latency distributed storage technology. It uses optical fiber as the storage medium and realizes data transmission and storage by reflecting signals. In FORM, each node has a fiber receiver and a fiber transmitter, which sends data to other nodes through optical fibers. Each node can read the optical signal and process it. Since the speed of light is very fast, FORM can achieve very low latency and very high data transmission rate. FORM technology is mainly used in military, aerospace, industrial automation and other fields to meet the needs of high-speed, high-reliability, and low-latency data transmission. It can be used in application scenarios such as real-time control systems, data acquisition systems, and distributed computing systems.

[0003] Dual Homed Ring is a network architecture that uses two ring structures, with each device connected to two different rings. When one of the rings fails, the other ring can continue to provide services, ensuring high availability and stability of the network. Dual Homed Ring is commonly used in key areas such as data centers, network switches, and servers. Compared with traditional redundant network architectures, dual homed rings have higher reliability and lower recovery time. At the same time, dual homed rings can also improve network throughput and load balancing capabilities by supporting multipath forwarding technology.

[0004] The dual redundant ring network structure used in the reflective memory network not only ensures the strong real-time and quasi-synchronous nature of the data, but also enables the entire ring to quickly perform redundant backup of the data in the system when one of the rings is damaged. Summary of the invention

[0005] The present invention aims to provide a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network, which can ensure strong real-time and quasi-synchronization of data while also ensuring data redundancy and fault tolerance through a dual redundant system.

[0006] To achieve the above object, the specific scheme of the present invention is as follows:

[0007] Step S101: inserting an optical fiber reflective memory card into each computer, the optical fiber reflective memory card adopts PCI, CPCI and PXI bus architecture, and connects each computer through optical fiber to form a ring network structure;

[0008] Step S102: Based on the ring network structure, each node in the ring network is reconnected by using optical fiber to form a one-main-one-backup network facility with the inner ring, forming a dual-redundant ring network structure;

[0009] Step S103: Connect the bridge nodes in each subsystem to two switches with the same structure to form a dual redundant structure, so as to achieve strong real-time and quasi-synchronization across regions;

[0010] Step S104: when sending and receiving data, the primary and backup systems simultaneously send and receive the same data, and the receiving end obtains the corresponding data by a de-redundancy method, thereby ensuring real-time backup and fault tolerance of the data;

[0011] Step S105: while sending and receiving data, the backup system monitors the primary system and uses a heartbeat detection mechanism to determine the data transmission status of the primary system. When it is determined that a failure occurs in the primary system link and causes transmission failure, it immediately switches to the backup system and organizes data retransmission;

[0012] Step S106: Perform fault detection on the system at each network port. If a switch fails, use a redundant switch to connect to each subsystem. If a loop fails, use a redundant loop to reconstruct the entire network.

[0013] Step S107: the standard clock source in the network shares the standard time with the reflective memory network, and the nodes in the same network read the standard time shared in the network by sending a request, and correct their own time according to the standard time.

[0014] Furthermore, in the dual redundant ring network topology structure described in step S102, a dual-port RM interface adapter is used on each computer device, the inner ring serves as the main system, and the outer ring serves as the backup system.

[0015] Furthermore, step S103 specifically includes:

[0016] Step S103.1, elect a node in the subsystem as a bridge node, the bridge node is connected to two switches through optical fibers, and a dual redundant switch is formed by connecting two switches and multiple bridge nodes. The two switches are connected to each other through redundant links, and the network devices in each ring network communicate through the two switches. When one of the switches or links fails, the other switch will automatically take over and maintain the normal operation of the network;

[0017] Step S103.2: configure the correct VLAN and network parameters on the switch, wherein the network parameters include IP address, subnet mask and default gateway.

[0018] Furthermore, step S104 specifically includes:

[0019] Step S104.1, the primary and backup systems simultaneously send the same data packet in the reverse direction, the primary system splits the data multiple times or uses different error correction coding methods to send multiple copies of the data on each channel or add redundant check information;

[0020] Step S104.2: After receiving the data on each channel, the receiving end starts to perform a de-redundant receiving operation, compares the data received on different channels, identifies and discards duplicate data, uses an error correction coding algorithm to decode and verify the received data, repairs existing errors or lost data, and performs data filtering and de-duplication operations based on redundant verification information or the characteristics of duplicate data, retaining only one complete copy of the data.

[0021] Furthermore, step S105 specifically includes:

[0022] Step S105.1, the primary system periodically sends a Keep-Alive heartbeat data packet to the backup system, wherein the heartbeat signal is a periodic network data packet;

[0023] Step S105.2, after receiving the heartbeat signal of the primary system, the backup system sends a response to the primary system, indicating that the primary system is in normal working condition. If the backup system does not receive the heartbeat signal from the primary system within a certain period of time, it is considered that the primary system has a fault or the network connection is interrupted, and takes corresponding measures to switch to the backup system;

[0024] Step S105.3: If a system failure or network connection interruption is detected and the sending end does not receive an ACK message from the receiving end indicating successful receipt of the data packet, the last data sending process failed and the system retransmits the data that failed to be sent.

[0025] Furthermore, step S106 specifically includes:

[0026] Step S106.1, performing quality assessment on the received optical signal and data flow at each network receiving port, detecting the noise level and distortion of the optical signal. If the optical signal quality is poor, it indicates that there is a loop fault;

[0027] Step S106.2: When it is determined that the system is abnormal and the cause is a switch failure, the redundant switch automatically triggers the switching of the standby switch, and the system switches the data traffic from the main switch to the redundant switch, and updates the network routing table and forwarding table;

[0028] Step S106.3: When it is determined that the system is abnormal and the reason is a loop failure, the system immediately triggers the redundant loop, and the interface adapter in the network node adopts a dual-port design to correctly transmit the data to the target node through another path in the ring network.

[0029] Furthermore, step S107 specifically includes:

[0030] Step S107.1, selecting a memory space in the reflective memory space as a unified time area, and dividing the unified time area into a standard time storage area and a time calibration request area;

[0031] Step S107.2: Select a satellite-borne high-precision clock source as the master clock of the entire system. The satellite-borne high-precision clock source is used as the global standard time. The standard time is published to the unified time zone in the reflective memory network. The time management computer in each subsystem reads the standard time and converts it into the system time of the local time zone.

[0032] Step S107.3, when other computers are ready to calibrate time, they send a time calibration request to the time management computer. The time management computer updates the status word of the time calibration request area. The status word of the time calibration request area is set to 0. The time management computer writes its latest time into the time storage area. When the data is written, the status word becomes 1. At this time, other computers read the time in the standard time storage area and calibrate it.

[0033] Compared with the existing reflective memory network time synchronization method, the present invention has the following advantages:

[0034] 1. The method of the present invention adopts a dual redundant ring network mechanism. Once the main ring link fails, the backup ring link will automatically take over to ensure the connectivity of the network. In contrast, the time synchronization method of the normal single-ring reflective memory network topology may not be able to cope with the impact of hardware failures and easily lead to system instability.

[0035] 2. The dual redundant switch design adopted by the method of the present invention has the ability to quickly respond to faults and can complete switching within milliseconds. The network equipment can hardly perceive the switching process, quickly respond to switch or link failures, and restore the network as soon as possible, thereby achieving seamless switching and continuous availability of network services.

[0036] 3. The method of the present invention adopts a heartbeat detection mechanism, and the backup system can monitor the operating status of the main system in real time. Once the main system fails or is disconnected, the backup system can immediately detect and take corresponding measures. The heartbeat detection mechanism can ensure that the status of the main system is accurately reflected in the backup system. If there is a problem with the main system, the backup system can quickly detect and take over the work, thereby ensuring the continuity and reliability of the system.

[0037] 4. The time synchronization method adopted by the method of the present invention uses a satellite-borne high-precision clock source as a standard clock, which can generate a stable and accurate time signal, thereby ensuring the time synchronization performance of the entire network. Newly connected nodes can calibrate their own time only by reading shared data, without the need for complex configuration or manual operation, thereby improving the flexibility and ease of use of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flow chart of a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network provided by an embodiment of the present invention;

[0039] Figure 2 A block diagram of a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network provided by an embodiment of the present invention;

[0040] Figure 3 The overall topological structure diagram provided by the embodiment of the present invention;

[0041] Figure 4 A dual redundant ring network fault-tolerant structure diagram provided by an embodiment of the present invention;

[0042] Figure 5 A structural diagram of a bridge node connected to a dual redundant switch provided by an embodiment of the present invention;

[0043] Figure 6 A flow chart of the redundant backup switching process method implemented by the present invention;

[0044] Figure 7 This is a flow chart of the time calibration algorithm implemented in the present invention. DETAILED DESCRIPTION

[0045] In order to make the features and advantages of the present invention more obvious and understandable, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Figure 1 This is a flow chart of a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network provided by an embodiment of the present invention. Figure 2 As shown, a block diagram of a multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network, the method comprising:

[0047] Step S101: insert a fiber optic reflective memory card into each computer. The fiber optic reflective memory card adopts a bus architecture such as PCI, CPCI, PXI, etc., and connects each computer through optical fiber to form a ring network structure. Each computer is equivalent to a node in the network.

[0048] Step S102: Based on the established ring network structure, each node in the ring network is reconnected using optical fibers to form a one-main-one-backup network facility with the inner ring, thereby forming a dual-redundant ring network structure;

[0049] Step S103: Connect the bridge nodes in each subsystem to two switches with the same structure to form a dual redundant structure, so as to achieve strong real-time and quasi-synchronization across regions;

[0050] Step S104: when sending and receiving data, the primary and backup systems simultaneously send and receive the same data, and the receiving end obtains the corresponding data by a de-redundancy method, thereby ensuring real-time backup and fault tolerance of the data;

[0051] Step S105: While sending and receiving data, the backup system monitors the primary system and uses a heartbeat detection mechanism to determine the data transmission status of the primary system. When it is determined that the transmission fails due to a failure in the primary system link, it immediately switches to the backup system and organizes data retransmission;

[0052] Step S106: Perform fault detection in the system at each network port; if the switch fails, use the redundant switch to connect to each subsystem; if the loop fails, use the redundant loop to reconstruct the entire network;

[0053] Step S107: The standard clock source in the network shares the standard time with the reflective memory network. The nodes in the same network read the standard time shared in the network by sending a request, and correct their own time according to the standard time.

[0054] Step S101 specifically includes:

[0055] First, insert the fiber optic reflective memory card into each computer, and connect the computers through the optical fiber to form a ring network;

[0056] The dual redundant ring network topology design in step S102 must first ensure that the network structures of the two loops are the same, and each node device connected by the optical fiber must be the same. These devices are connected to each other through ports, and the ports on the devices use dual-port RM interface adapters. Data packets are transmitted in reverse between the two rings at the same time. If one of the rings fails, the data packet can be transmitted through the other ring to ensure the connectivity of the network. When designing a dual redundant ring network structure, the following points need to be noted: There must be at least two connection points between the two rings to ensure the reliability of the network. There must be a management node between the two rings, which is responsible for monitoring and managing the status of the two rings and automatically switching when a failure occurs. The network equipment needs to be reasonably configured to avoid single point failures. Step S102 specifically includes:

[0057] Step 1: Build Figure 3The overall system topology shown in the figure uses optical fibers to connect the already connected switches and nodes, and the nodes and nodes again to form a dual redundant ring network structure.

[0058] Step 2: If Figure 4 As shown, in each subsystem of the ring network, two optical fibers are used to connect the ports of each node device, and the two loops can perform bidirectional data transmission.

[0059] The dual redundant switch in step S103 is connected to the bridge node. The dual redundant switch is usually composed of two switches and multiple bridge nodes connected, wherein the two switches are connected to each other through redundant links, and the network devices in each ring network communicate through the two switches. When one of the switches or links fails, the other switch will automatically take over and maintain the normal operation of the network. Configure the correct VLAN (virtual local area network) and other network parameters on the switch to ensure the correct communication between the bridge node and other network devices. Including setting the correct IP address, subnet mask, default gateway, etc. The dual redundant switch uses two switches for redundant design. Once the main switch fails, the backup switch will automatically take over, thereby avoiding single point failure. The switching time of the dual redundant switch is usually at the millisecond level, which can quickly respond to failures and restore the network as soon as possible. Through the dual redundant design, the dual redundant switch can improve the reliability and stability of the network. On the basis of the redundant dual ring network, information sharing between multiple rings is realized through the bridge node. The bridge node adopts a four-port RM interface adapter. First, the bridge node in each ring network is connected to the two nodes in the ring network and the switch outside the ring network. Each bridge node is connected to two switches, one is the main switch and the other is the backup switch, forming a dual redundant switch structure. When one of the switches is damaged, the ring network can use the undamaged switch to ensure normal communication between the subsystems. At the same time, the dual redundant switch can quickly remove the damaged nodes in the network and reconstruct the network.

[0060] Step S103 specifically includes:

[0061] Step 1: Elect a node in the subsystem as a bridge node. The bridge node is connected to two switches through optical fiber to form a Figure 5 The topology shown.

[0062] Step 2: Configure the correct VLAN (Virtual Local Area Network) and other network parameters on the switch to ensure correct communication between the bridge node and other network devices, including setting the correct IP address, subnet mask, default gateway, etc.

[0063] In step S104, the main system and the backup system work simultaneously in the network, and have the functions of multi-redundant sending and de-redundant receiving data. Multi-redundant sending: During the data transmission process, the main switch and the backup switch can send data to multiple target nodes at the same time. De-redundant receiving: At the data receiving end, multiple backup nodes can be set to receive data. It is designed to send data with exactly the same effective load on both the main switch and the backup switch channels when sending data; for the network terminal receiving data, it is responsible for receiving the same copy of the data on the two channels of the main switch and the backup switch, and automatically performing data de-redundancy operations at the receiving site. The overall process avoids the impact of failures that may occur due to single-channel transmission on transmission efficiency and reliability. Step S104 specifically includes:

[0064] Step 1: Both systems send the same data packet in reverse at the same time. In order to increase the redundancy of the data, the main system adopts different strategies, splits the data multiple times or uses different error correction coding methods, sends multiple copies of the data on each channel or adds redundant check information.

[0065] Step 2: After receiving the data on each channel, the receiver starts to perform redundancy reception operations. Compare the data received on different channels, identify and discard duplicate data. Use the error correction coding algorithm to decode and verify the received data to repair possible errors or lost data. According to the redundant check information or the characteristics of the duplicate data, perform data filtering and deduplication operations to retain only one complete copy of the data.

[0066] In step S105, the backup system is monitored through heartbeat detection, and the main system and the backup system establish a TCP connection. After the connection is established, the monitoring mechanism can be realized with each other. After the TCP connection is established, the Keep-Alive option of TCP can be used to enable the heartbeat detection function. The main system sends a small Keep-Alive data packet to the backup system at regular intervals. If the backup system does not receive the heartbeat packet from the main system within the timeout period, and the sending end does not receive the ACK message returned by the receiving end, the backup system will think that the main system may have a problem and the data transmission is unsuccessful. At this time, the backup system can be immediately enabled and restarted, and the retransmission mechanism can be started to retransmit the data sent last time. Step S105 specifically includes:

[0067] Step 1: The primary system will periodically send Keep-Alive heartbeat packets to the backup system. These heartbeat signals are periodic network packets.

[0068] Step 2: After receiving the heartbeat signal from the primary system, the backup system sends a response to the primary system, indicating that the primary system is in normal working condition. If the backup system does not receive the heartbeat signal from the primary system within a certain period of time, it will be considered that the primary system has failed or the network connection is interrupted, and take corresponding measures to switch to the backup system.

[0069] Step 3: If the sender does not receive the ACK message from the receiver indicating that the data packet has been successfully received after a system failure or network connection is detected, the last data transmission process has failed and the system needs to retransmit the data that failed to be sent. This includes resending lost or damaged data packets or resending the entire data frame. At the same time, before retransmitting data, it is necessary to wait for an appropriate time window to ensure that there is an available network link to avoid continuous transmission failures.

[0070] In step S106, redundant switches, namely, a main switch and a backup switch, are used in the network topology. The backup switch is in a standby state and monitors the status of the main switch at all times. When the main switch fails, the backup switch can automatically trigger the switching of the backup switch and immediately take over the network data transmission work to ensure the connectivity of the network. Step S106 specifically includes:

[0071] Step 1: Evaluate the quality of the received optical signal and data traffic at each network receiving port, and detect the noise level and distortion of the optical signal. If the optical signal quality is poor, it indicates a loop fault.

[0072] Step 2: When the system is judged to be abnormal and the cause is a switch failure, the redundant switch can automatically trigger the switch of the backup switch. The system will switch the data traffic from the main switch to the redundant switch. And update the network routing table, forwarding table and other configurations to ensure that data can be transmitted normally.

[0073] Step 3: If Figure 6 As shown in the figure, when the system is judged to be abnormal and the reason is a loop failure, the system immediately triggers the redundant loop. Among them, the interface adapter in the network node adopts a dual-port design, and the redundant loop design correctly transmits the data to the target node through another path in the ring network, with good fault tolerance and reconstruction capabilities. There are two situations for loop failure: when a node is detected to be damaged, the damaged node is automatically removed and the network is rebuilt; when an optical cable failure is detected, the faulty optical cable is automatically bypassed, and the ring network is rebuilt through the remaining optical cables in the inner ring and outer ring to ensure that all nodes communicate normally. The switching process needs to ensure the continuity and consistency of the data to avoid data loss or repeated transmission.

[0074] In step S107, firstly, a satellite-borne high-precision clock source is selected as the master clock of the entire system. When working normally, it can generate a data packet of standard time. After receiving the standard time data packet, the time management computer of each subsystem converts it into the standard time of the time zone and publishes the standard time in the reflective memory network. If the standard clock source is damaged, the time management computer of each subsystem will serve as a backup clock source and continue to publish the standard time to the network. Step S107 specifically includes:

[0075] Step 1: Time synchronization algorithm such as Figure 7 As shown, a section of memory space is selected in the reflective memory space as the time unified area. The time unified area is divided into a standard time storage area and a time calibration request area.

[0076] Step 2: Select a satellite-borne high-precision clock source as the master clock for the entire system. Use it as the global standard time and publish the standard time to the time zone in the reflective memory network. The time management computer in each subsystem reads the standard time and converts it into the system time of the local time zone.

[0077] Step 3: When other computers want to calibrate time, they send a calibration request to the time management computer. The time management computer will update the status word of the calibration request area, and the status word of the calibration request area will be set to 0. At this time, the time management computer will write its latest time into the time storage area. When the data is written, the status word becomes 1, and other computers read the time in the standard time storage area and calibrate it.

[0078] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-ring topology redundancy and data strong real-time fault tolerance method for an optical fiber reflective memory network, characterized in that: The following steps are involved: Step S101: inserting an optical fiber reflective memory card into each computer, the optical fiber reflective memory card adopts PCI, CPCI and PXI bus architecture, and connects each computer through optical fiber to form a ring network structure; Step S102: Based on the ring network structure, each node in the ring network is reconnected by using optical fiber to form a one-main-one-backup network facility with the inner ring, forming a dual-redundant ring network structure; Step S103: Connect the bridge nodes in each subsystem to two switches with the same structure to form a dual redundant structure, so as to achieve strong real-time and quasi-synchronization across regions; Step S104: when sending and receiving data, the primary and backup systems simultaneously send and receive the same data, and the receiving end obtains the corresponding data by a de-redundancy method, thereby ensuring real-time backup and fault tolerance of the data; Step S105: while sending and receiving data, the backup system monitors the primary system and uses a heartbeat detection mechanism to determine the data transmission status of the primary system. When it is determined that a failure occurs in the primary system link and causes transmission failure, it immediately switches to the backup system and organizes data retransmission; Step S106: Perform fault detection on the system at each network port. If a switch fails, use a redundant switch to connect to each subsystem. If a loop fails, use a redundant loop to reconstruct the entire network. Step S107: the standard clock source in the network shares the standard time with the reflective memory network, and the nodes in the same network read the standard time shared in the network by sending a request, and correct their own time according to the standard time.

2. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 1, characterized in that: In the dual redundant ring network topology structure described in step S102, a dual-port RM interface adapter is used on each computer device, the inner ring is used as the main system, and the outer ring is used as the backup system.

3. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 2, characterized in that: Step S103 specifically includes: Step S103.1, elect a node in the subsystem as a bridge node, the bridge node is connected to two switches through optical fibers, and a dual redundant switch is formed by connecting two switches and multiple bridge nodes. The two switches are connected to each other through redundant links, and the network devices in each ring network communicate through the two switches. When one of the switches or links fails, the other switch will automatically take over and maintain the normal operation of the network; Step S103.2: configure the correct VLAN and network parameters on the switch, wherein the network parameters include IP address, subnet mask and default gateway.

4. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 3 is characterized in that: Step S104 specifically includes: Step S104.1, the primary and backup systems simultaneously send the same data packet in the reverse direction, the primary system splits the data multiple times or uses different error correction coding methods to send multiple copies of the data on each channel or add redundant check information; Step S104.2: After receiving the data on each channel, the receiving end starts to perform a de-redundant receiving operation, compares the data received on different channels, identifies and discards duplicate data, uses an error correction coding algorithm to decode and verify the received data, repairs existing errors or lost data, and performs data filtering and de-duplication operations based on redundant verification information or the characteristics of duplicate data, retaining only one complete copy of the data.

5. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 4, characterized in that: Step S105 specifically includes: Step S105.1, the primary system periodically sends a Keep-Alive heartbeat data packet to the backup system, wherein the heartbeat signal is a periodic network data packet; Step S105.2, after receiving the heartbeat signal of the primary system, the backup system sends a response to the primary system, indicating that the primary system is in normal working condition. If the backup system does not receive the heartbeat signal from the primary system within a certain period of time, it is considered that the primary system has a fault or the network connection is interrupted, and takes corresponding measures to switch to the backup system; Step S105.3: If a system failure or network connection interruption is detected and the sending end does not receive an ACK message from the receiving end indicating successful receipt of the data packet, the last data sending process failed and the system retransmits the data that failed to be sent.

6. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 5, characterized in that: Step S106 specifically includes: Step S106.1, performing quality assessment on the received optical signal and data flow at each network receiving port, detecting the noise level and distortion of the optical signal. If the optical signal quality is poor, it indicates that there is a loop fault; Step S106.2: When it is determined that the system is abnormal and the cause is a switch failure, the redundant switch automatically triggers the switching of the standby switch, and the system switches the data traffic from the main switch to the redundant switch, and updates the network routing table and forwarding table; Step S106.3: When it is determined that the system is abnormal and the reason is a loop failure, the system immediately triggers the redundant loop, and the interface adapter in the network node adopts a dual-port design to correctly transmit the data to the target node through another path in the ring network.

7. The multi-ring topology redundancy and data strong real-time fault tolerance method of a fiber optic reflective memory network according to claim 6, characterized in that: Step S107 specifically includes: Step S107.1, selecting a memory space in the reflective memory space as a unified time area, and dividing the unified time area into a standard time storage area and a time calibration request area; Step S107.2: Select a satellite-borne high-precision clock source as the master clock of the entire system. The satellite-borne high-precision clock source is used as the global standard time. The standard time is published to the unified time zone in the reflective memory network. The time management computer in each subsystem reads the standard time and converts it into the system time of the local time zone. Step S107.3, when other computers are ready to calibrate time, they send a time calibration request to the time management computer. The time management computer updates the status word of the time calibration request area. The status word of the time calibration request area is set to 0. The time management computer writes its latest time into the time storage area. When the data is written, the status word becomes 1. At this time, other computers read the time in the standard time storage area and calibrate it.

Images