Point-to-point airborne front-end time synchronization method

Abstract

The point-to-point airborne front-end time synchronization method of the application is suitable for time synchronization between independent FC nodes in a communication network without passing through a switch, and the method comprises the following steps: determining a point-to-point clock server and a point-to-point clock client according to an upper-layer application; after power-on connection of a node, periodically sending local clock or clock information obtained from an external channel to the point-to-point clock client by the point-to-point clock server; after receiving the clock information, calculating a difference between a system RTC and a local RTC of the client by the point-to-point clock client; and obtaining a calendar time reference sent by the point-to-point clock server, and replacing a calendar time reference of the point-to-point clock client with the calendar time reference, so as to solve a time synchronization problem in a point-to-point connection topology of an FC network front end.

Description

Technical Field

[0001] This invention belongs to the technical field of network technology, and in particular relates to a point-to-point airborne front-end time synchronization method. Background Technology

[0002] Fiber Channel (FC) is widely used in network storage and switching due to its high bandwidth, low latency, and support for multiple upper-layer protocols. Globally or locally consistent time is one of the fundamental services that FC networks provide to upper-layer applications.

[0003] Existing FC networks have implemented network clock synchronization based on the system RTC (Real Time Clock) within the switching network. The clock server uses the clock synchronization primitives in the FC protocol as a carrier, which are forwarded to each device node in the switching network via the switch. At the same time, all nodes in the switching network obtain the same system RTC value, and all nodes maintain dynamic consistency with the same reference clock, ensuring that all nodes in the switching network work collaboratively on the same time plane. However, the system RTC only solves the clock synchronization problem in FC switching networks. In FC networks, in addition to the switching network nodes, there are also some FC nodes that use a point-to-point connection, such as remote radar nodes and remote imaging nodes. Since these nodes are not directly connected to the switch through the FC network, the clock synchronization primitives cannot arrive in real time, becoming clock synchronization islands. Therefore, solving the clock synchronization problem of point-to-point connected front-end nodes and ensuring that the front-end and the switching network are in sync is of great significance for ensuring task execution.

[0004] In view of this, the present invention is hereby proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a point-to-point airborne front-end time synchronization method to solve the time synchronization problem in point-to-point connection topologies of FC network front-ends. This method expands the time synchronization range of the switching topology to the entire network node, achieving time synchronization across the entire task domain and providing more efficient and reliable services for upper-layer applications. The technical solution of this invention has many beneficial effects, as described below:

[0006] A point-to-point airborne front-end time synchronization method is provided, applicable to time synchronization between independent FC nodes in a communication network without going through a switch. The method comprises:

[0007] Based on the upper-layer application, the dot-matrix clock server and dot-matrix clock client are determined. After the node powers on and connects, the dot-matrix clock server periodically sends its local clock or clock information obtained from external channels to the dot-matrix clock client. Upon receiving the information, the clock client calculates the difference between the system RTC and the client's local RTC, and obtains the calendar time base sent by the dot-matrix clock server, replacing its own calendar time base with the calendar time base.

[0008] At any time or within a preset time period, the point clock client responds to an externally transmitted signal to determine the RTC of the system at the current time. The RTC of the system at the current time is determined based on the difference and the local RTC of the point clock client.

[0009] At any given time or within a preset time period, the point clock client responds to an externally transmitted signal to determine the current calendar time, which is determined based on the system's RTC and the client's local time reference.

[0010] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0011] This invention addresses the issue of FC nodes used in a point-to-point connection manner, such as remote radar nodes or remote imaging nodes, avoiding the situation where clock synchronization primitives cannot arrive in real time and become clock synchronization islands because these nodes are not directly connected to the switch through the FC network. It expands the time synchronization range of the switching topology to the entire network node, realizing time synchronization of the entire task domain and providing more efficient and reliable services for upper-layer applications. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A schematic diagram of point-to-point direct clock synchronization;

[0014] Figure 2 A schematic diagram of point-to-point connection clock synchronization involving an exchange network. Detailed Implementation

[0015] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0016] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this invention, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0017] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0018] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that aspects can be practiced without these specific details. To enable those skilled in the art to better understand the invention, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined as "first" and "second" may explicitly or implicitly include one or more of that feature. In the description of the invention, unless otherwise stated, "a plurality of" means two or more.

[0019] like Figure 1The point-to-point airborne front-end time synchronization method shown is applicable to time synchronization between independent FC nodes in a communication network that do not go through a switch. It enables time synchronization between two nodes with independent communication capabilities, solves the clock synchronization problem of point-to-point connected front-end nodes, and ensures time consistency between the front-end and the switching network. The method includes:

[0020] Based on the upper-layer application, the dot-matrix clock server and dot-matrix clock client are determined. After the nodes are powered on and connected, the dot-matrix clock server periodically sends clock information from its local clock (without a switching network) or from an external channel (with a switching network) to the dot-matrix clock client. Upon receiving the information, the clock client calculates the difference between the system's RTC and the client's local RTC, and obtains the calendar time base sent by the dot-matrix clock server, replacing its own calendar time base with this new calendar time base.

[0021] At any given time or within a preset time period, the point-to-point clock client responds to externally transmitted signals to determine the system's current RTC. The current system RTC is determined based on the difference between the system's RTC and the point-to-point clock client's local RTC. Preferably, the point-to-point clock server's RTC synchronization is achieved through time synchronization primitives, and calendar time synchronization is achieved through ELS frame communication. Clock synchronization primitives refer to the FC protocol primitives carrying RTC information that are logically sent out after a node is initialized as a clock server.

[0022] At any given time or within a preset time period, the point clock client responds to externally transmitted signals to determine the current calendar time, which is determined based on the system's RTC and the client's local time base.

[0023] After completing the above two steps, clock synchronization is achieved for the point-to-point connection, that is, time synchronization between the server and the client.

[0024] As a specific implementation method provided in this case, after the ELS frame is powered on, the point clock server periodically sends out network probe frames. The content of the network probe frames includes the node's own port information, local time, locally stored netlist and synchronization period, and is used to implement network management.

[0025] After receiving the time synchronization primitive and the network probe frame, the DotClock client enables internet access. The DotClock server sends the obtained data to the system RTC for clock calibration, and updates the local time information and network management information according to the obtained calendar time reference information to achieve clock synchronization. Furthermore, it monitors the DotClock server based on the periodic information carried in the network probe frame. If no network probe frame is received within a timeout period, the DotClock client disconnects from the network and ends clock synchronization.

[0026] Furthermore, each independent FC node is an independently communicating logical node. One or more logical nodes exist simultaneously on a physical device, and each logical node communicates with the switching network or with nodes of external communication devices in a point-to-point manner.

[0027] As a specific implementation method provided in this case, the dot-matrix clock server and dot-matrix clock client are determined according to the upper-layer application, including:

[0028] If there is an external network, the node that communicates with the external network will be the point-to-point clock server, and the node connected to the point-to-point clock server will be the point-to-point clock client.

[0029] In the absence of an external network, any independent FC node acts as a dot-to-dot clock server, and the FC node connected to the dot-to-dot clock server acts as a dot-to-dot clock client.

[0030] The node that actively initiates clock synchronization is called the point-to-point clock server, and the node that passively receives synchronization is called the point-to-point clock client. If no switching network exists, either of the two directly connected nodes can serve as the point-to-point clock server; if a switching network exists, the logical node that coexists with the switching network node on the same physical device is generally used as the point-to-point clock server; the logical node located at the front end, existing independently, and far from the switching network is used as the point-to-point clock client.

[0031] Furthermore, external channels refer to internal resources or interfaces that allow information exchange between two logical nodes within the same physical device, wherein:

[0032] Within the same physical device, the node directly connected to the switching network is called the switching network logical node C, which acts as the switching network clock synchronization client in the switching network synchronization. The node on the same physical device as the switching network logical node C is called the point-to-point connection logical node A, and the front-end node of the point-to-point connection logical node A is called the point-to-point clock client B. That is, node A acts as the point-to-point clock server in the point-to-point clock synchronization.

[0033] The switching network logical node C and the point-to-point connection logical node A exchange clock synchronization information through internal dual ports or shared memory, and the exchange information is notified via message in the form of shared dual port interrupt.

[0034] Point-to-point connection logical node A obtains system RTC and time information from its internal dual-port or shared memory. Switching network logical node C periodically obtains system RTC and time information from the switching network. Switching network logical node C and point-to-point connection logical node A share the same hardware crystal oscillator, and their local RTCs are identical. Therefore, their time-synchronized system RTCs are identical. When the switching network logical node completes switching network system RTC synchronization, point-to-point connection logical node A directly obtains the system RTC value acquired by switching network logical node C through its internal dual-port or shared memory.

[0035] Point-to-point connection logical node A acts as a point-to-point clock server, communicating with point-to-point clock clients. It periodically obtains the system RTC value from the switching network logical node C through internal dual-port or shared memory and sends it to the point-to-point clock clients via clock primitives to achieve full network RTC synchronization. That is, C gets data from the switching network, A gets data from C, B gets data from A, and so on, achieving full network RTC synchronization of A, B, V, and the switching network.

[0036] When the switching network logical node C completes the switching network system time synchronization, it updates the switching network system calendar time base to the shared dual-port memory in real time according to the switching network synchronization cycle, and notifies the point-to-point connection logical node A through an interrupt. Point-to-point connection logical node A, as the point-to-point clock server, periodically sends the calendar time base to the point-to-point clock client b through ELS frames to achieve calendar time synchronization of the entire network.

[0037] Working principle:

[0038] When two nodes are completely independently connected, as shown in the attached figure. Figure 2 As shown, Node A is initialized as a point-to-point clock server by the upper-layer application, and Node B is initialized as a point-to-point clock client by the application. After power-on, Node A uses its local RTC as the system RTC, automatically sends clock synchronization primitives outward, and combines its own node information, time, and netlist information into a network probe frame. After receiving the above primitives and network probe frames, Node B realizes clock synchronization by calculating the difference between the system RTC and the local RTC, updates its local time and local netlist using the difference and time information, and monitors the point-to-point clock server according to the period of the network probe frames.

[0039] When point-to-point connections and switched connections intersect, such as Figure 2As shown, node C is a switching node, and nodes A and C reside on the same physical device. Nodes A and C can exchange information via an internal dual-port. Node B is a front-end node. Node C participates in the switching network clock synchronization, periodically obtaining the system RTC value, time information, and network list from the switch. While performing switching clock synchronization, node A uses internal resources to calculate the difference between the system RTC and the local RTC to synchronize the clock and inform node A. Node A is initialized as a point-to-point clock server by the upper-layer application, and node B is initialized as a point-to-point clock client. Node A uses the switching system RTC obtained from node C as the point-to-point synchronization system RTC, sending clock synchronization primitives outwards. It then uses the obtained time information and network list to organize network probe frames for outward transmission. After receiving the primitives and probe frames, node B achieves clock synchronization by calculating the difference between the system RTC and the local RTC, updates its local time and network list using the difference and time information, and monitors the point-to-point clock server according to the network probe frame period. The time synchronization of A, B, and C achieves clock synchronization across the entire task domain.

[0040] Example

[0041] This invention is applicable to clock synchronization between two FC nodes with independent communication capabilities in a directly connected topology. The synchronization mechanism is as follows:

[0042] 1) such as Figure 1As shown, the nodes are completely independently connected. Both nodes have a local RTC, which is essentially a timer with a resolution of 100ns and a data width of 42 bits, which automatically accumulates after power-on. The local RTC is driven by the node's own crystal oscillator at the underlying hardware level. Local time is calculated by counting the number of oscillation ticks of the crystal oscillator and multiplying it by the duration of a single tick. After the directly connected nodes A and B are powered on, there is a fixed initial tick difference Δ0 = TA - local(0) - TB - local(0) between them due to the different startup times. When the upper-layer application sets node A as the clock server and node B as the clock client, node A uses its current RTC as the system RTC and sends FC protocol time synchronization primitives outward according to the time synchronization period set by the application. The primitives arrive at node B after transmitting Δ1 time. At sampling point t, the local RTC time of node B is TB-local(t). Therefore, the system RTC value obtained by node B through calculation is TB-sys(t) = TB-local(t) + Δ0 + Δ1. When the upper-layer application sets the year, month, day, and time for node A, node A sends the year, month, day, time, and node A's synchronization period outward through FC-ELS messages. After receiving them, node B performs time conversion on the calculated system RTC (TB-sys(t)) and adds the conversion result to node A's year, month, day, and time to achieve system time synchronization. In addition, node B will perform synchronization monitoring settings according to node A's synchronization period. If no synchronization primitive or synchronization ELS frame is received for more than 5 monitoring periods, synchronization is considered to have failed.

[0043] 2) such as Figure 2 As shown, nodes A and C are two logical nodes within the same module, driven by the same crystal oscillator at the underlying hardware level. Therefore, their local RTCs are completely identical. Node C is directly connected to the switch, participating in the switching network clock synchronization and obtaining the switching network's date, time, and system RTC. Node C and node A share a dual-port memory within the same module. After node C completes a switching network clock synchronization, it writes the switching network's date, time, and system RTC into the shared dual-port memory and notifies node A via an internal interrupt. Node A retrieves the information from the shared dual-port memory after the interrupt is triggered. The process of node C writing to node A and retrieving the information is very short and negligible. Since node A and node C have the same local RTC, node A does not need to recalculate. Instead, it uses the currently acquired switching network system RTC as the point-to-point connection system RTC and sends the FC protocol time synchronization primitive outward according to the clock synchronization period set by the application. The primitive arrives at node B after a transmission time of Δ1. At sampling point t, node B's local RTC time is TB-local(t). Therefore, the system RTC value obtained by node B through calculation is TB-sys(t) = TB-local(t) + Δ1 + Δ0, where:

[0044] △0 represents the initial TICK difference between the crystal oscillator of the module where node B is located and the crystal oscillators of the modules where nodes A and C are located. △0 = TB-local(0) - TAC-local(0). Besides the clock being the same as the primitive, node A sends the year, month, day, and time information obtained from node C outwards via FC-ELS messages. After receiving this, node B performs time conversion on the calculated system RTC and adds the conversion result to the year, month, day, and time of node A to achieve system time synchronization. Furthermore, node B will perform synchronization monitoring settings according to the synchronization cycle of node A. If no synchronization primitive or synchronization ELS frame is received for more than 5 monitoring cycles, synchronization is considered to have failed. Figure 2 In the point-to-point clock synchronization process shown, since the system RTC and date / time of node A are sourced from node C of the switching network, clock synchronization of the entire task domain of the ABC switching network is ultimately achieved, providing a reliable guarantee for the time consistency of the entire task domain.

[0045] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the core ideas of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made to the invention without departing from the principles of the invention, and these improvements and modifications also fall within the protection scope of the invention claims.

Claims

1. A point-to-point airborne front-end time synchronization method, applicable to time synchronization between independent FC nodes in a communication network without going through a switch, characterized in that... The methods include: Based on the upper-layer application, the dot-matrix clock server and dot-matrix clock client are determined. After the node powers on, the dot-matrix clock server periodically sends local clock information or clock information obtained from external channels to the dot-matrix clock client. Upon receiving the information, the clock client calculates the difference between the system RTC and the client's local RTC, and obtains the calendar time reference sent by the dot-matrix clock server, replacing its own calendar time reference with the calendar time reference. At any given time or within a preset time period, the point-to-point clock client responds to an externally transmitted signal to determine the system's current RTC. The current RTC is determined based on the difference and the point-to-point clock client's local RTC. At any given time or within a preset time period, the point clock client responds to an externally transmitted signal to determine the current calendar time, which is determined based on the system's RTC and the client's local time reference. Among them, the RTC synchronization of the dot clock server is transmitted through time synchronization primitives, and the calendar time synchronization is transmitted through ELS frame communication. After the ELS frame is powered on, the point clock server periodically sends out network probe frames. The content of the network probe frames includes the node's own port information, local time, locally stored netlist and synchronization period, and is used to implement network management. After receiving the time synchronization primitive and the network probe frame, the Dot Clock client enables internet access. The Dot Clock server sends the obtained data to the system RTC for clock calibration, and updates the local time information and network management information according to the acquired calendar time reference information to achieve clock synchronization. Furthermore, the client monitors the Dot Clock server based on the periodic information carried in the network probe frame. If no network probe frame is received within a timeout period, the Dot Clock client disconnects from the internet and ends clock synchronization.

2. The point-to-point airborne front-end time synchronization method according to claim 1, characterized in that, Each of the independent FC nodes is an independently communicating logical node. One or more logical nodes exist simultaneously on a physical device. Each of the logical nodes communicates with the switching network or is connected to the nodes of external communication devices in a point-to-point manner.

3. The point-to-point airborne front-end time synchronization method according to claim 1, characterized in that, The dot-matrix clock server and dot-matrix clock client are determined based on the upper-layer application, including: If an external network is available, the node communicating with the external network will be designated as the point-to-point clock server, and the node connected to the point-to-point clock server will be designated as the point-to-point clock client. In the absence of an external network, any independent FC node acts as a dot-to-dot clock server, and the FC node connected to the dot-to-dot clock server acts as a dot-to-dot clock client.

4. The point-to-point airborne front-end time synchronization method according to claim 3, characterized in that, External channels refer to internal resources or interfaces that allow information exchange between two logical nodes within the same physical device, where: Within the same physical device, the node directly connected to the switching network is designated as the switching network logical node C, and acts as the switching network clock synchronization client during switching network synchronization; the node on the same physical device as the switching network logical node C is designated as the point-to-point connection logical node A, and the front-end node of the point-to-point connection logical node A acts as the point-to-point clock client B. The switching network logical node C and the point-to-point connection logical node A exchange clock synchronization information through internal dual ports or shared memory, and the exchange information is notified via message in the form of shared dual port interrupt. The point-to-point connection logical node A obtains system RTC and time information from its internal dual-port or shared memory, while the switching network logical node C periodically obtains system RTC and time information from the switching network.

5. The point-to-point airborne front-end time synchronization method according to claim 4, characterized in that, The switching network logical node C and the point-to-point connection logical node A share the same hardware crystal oscillator. When a switching network logical node completes the switching network system RTC synchronization, point-to-point connection logical node A directly obtains the system RTC value obtained by switching network logical node C through its internal dual-port or shared memory; The point-to-point connection logic node A acts as a point-to-point clock server, communicating with the point-to-point clock client. It periodically sends the system RTC value obtained by the internal dual-port or shared memory acquisition and switching network logic node C to the point-to-point clock client through clock primitives to achieve full network RTC synchronization. When the switching network logical node C completes the switching network system time synchronization, it updates the switching network system calendar time base to the shared dual-port memory in real time according to the synchronization cycle of the switching network, and notifies the point-to-point connection logical node A through an interrupt. Point-to-point connection logical node A, as the point-to-point clock server, periodically sends the calendar time base to the point-to-point clock client B through ELS frames to achieve calendar time synchronization of the entire network.

Images