RapidIO Network Management Method for Signal Processing System

Abstract

The present invention provides a rapidIO network management method, aiming to provide a method for reducing software operation anomalies caused by network congestion. This solution is achieved through the following technical solutions: Each processor node on the signal processing module is interconnected through a rapidIO network switch to form a rapidIO sub-network, and the rapidIO sub-networks on each module are interconnected to form a rapidIO backbone network. The master node in the sub-network maintains the status of each node in the sub-network, executes fault handling measures when an anomaly occurs, and periodically sends the sub-network status to the master node of the main module. The master node of the main module maintains the status of each sub-network. When the master node of the main module fails, the master node of the backup module maintains the status of the backbone network.

Description

Technical Field

[0001] The present invention belongs to the technical field of network communication, and in particular relates to a rapidIO network management method for a signal processing system. Background Art

[0002] With the development of the signal processing field towards the direction of high speed and high efficiency, the computing resource nodes inside the signal processing system increase exponentially with the increase of the operation scale. The interconnection communication between large-scale computing resources has become a key factor affecting the performance of the signal processing system. As an interconnection standard in the embedded system, the rapidIO bus is suitable for the tightly coupled working environment of multiple computing resource nodes and supports hot plugging of each node. In response to the requirement of dividing various functions into field-replaceable unit modules in the signal processing system, the rapidIO bus can obtain higher system-level performance and support dynamic networking of each unit module. In the signal processing system, a large number of DSP, CPU, and FPGA devices are interconnected by rapidIO, which can quickly and efficiently complete the data transfer function. However, in practice, when the network management node goes offline, there is no management node in the network, resulting in the inability of each node in the network to obtain the status of the peer node, leading to data transmission failure, causing data congestion, and further causing the processor node to malfunction. For example, if a receiving node in the network goes offline and there is no message notification, the data sender continues to send, resulting in processor communication congestion and reduced data interaction efficiency. Summary of the Invention

[0003] In view of this, the embodiments of the present disclosure provide a rapidIO network management method for a signal processing system, which improves the system scalability and adaptability, and solves the problem that when the network management node goes offline, there is no management node in the network, resulting in the inability of each node in the network to obtain the status of the peer node, leading to data transmission failure, causing data congestion, and further causing the processor node to malfunction.

[0004] A rapidIO network management method for a signal processing system, including a signal processing main module, a signal processing standby module, and a plurality of signal processing slave modules set in a dual-redundancy manner. The signal processing standby module is a hot backup of the signal processing main module. When the signal processing main module goes offline, the signal processing standby module goes online;

[0005] The signal processing main module includes a first switch, a first main processing node CPU, and a plurality of first sub-processing nodes. The first main processing node CPU obtains the status information of the plurality of first sub-processing nodes through the first switch and generates a first information table;

[0006] The signal processing standby module includes a second switch, a second main processing node CPU, and multiple second sub-processing nodes. The second main processing node CPU obtains the status information of the multiple second sub-processing nodes through the second switch and generates a second information table.

[0007] Each signal processing slave module includes a third switch, a third main processing node CPU, and multiple third sub-processing nodes. The third main processing node CPU obtains the status information of the multiple third sub-processing nodes through the third switch and generates a subnet information table.

[0008] The first main processing node CPU and the second main processing node CPU obtain the subnet information tables generated by all the signal processing slave modules, and respectively summarize them with their own first information table or second information table into a backbone network information table. The backbone network information table is read in real time by all the signal processing slave modules, knows the working status of each node of all other signal processing slave modules and the signal processing main module, and determines whether to perform data interaction.

[0009] When the first main processing node CPU of the signal processing main module fails, the second main processing node CPU of the signal processing standby module manages the backbone network and solves the communication congestion problem caused by node offline.

[0010] Beneficial effects:

[0011] That is, the signal processing system is divided into rapidIO subnets according to the signal processing modules. Each rapidIO subnet jointly forms a rapidIO backbone network. The status of each subnet is maintained by the main node of the slave module, and the backbone network is maintained by the main node of the main module. When the link between the subnet and the backbone network is abnormal, the subnet can work normally under the management of the main node of the slave module without affecting the work of the backbone network. At the same time, the main node of the slave module obtains the status of the subnet cascade port by means of trial access, avoiding processor communication congestion caused by data access after the subnet cascade port fails. If the receiving node goes offline and there is no message notification, the data sender continues to send, resulting in processor communication congestion. Description of the Drawings

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained according to these drawings without creative efforts.

[0013] Figure 1 It is a schematic diagram of the network structure;

[0014] Figure 2 It is a schematic diagram of the structure of the signal processing main module;

[0015] Figure 3 Schematic structural diagram of the signal processing standby module;

[0016] Figure 4 Schematic structural diagram of the slave module. Specific implementation manners

[0017] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0018] The following uses specific specific examples to illustrate the implementation manners of the present disclosure. Those skilled in the art can easily understand other advantages and effects of the present disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. The present disclosure can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present disclosure without making creative efforts belong to the scope of protection of the present disclosure.

[0019] It should be noted that the following describes various aspects of the embodiments within the scope of the appended claims. It should be obvious that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is illustrative only. Based on the present disclosure, those skilled in the art should 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 described herein can be used to implement the device and / or practice the method. In addition, this device and / or practice this method can be implemented using other structures and / or functions in addition to one or more of the aspects described herein.

[0020] Such as Figure 1 shown, the rapidIO network management method of the signal processing system, the signal processing main module, the signal processing standby module and multiple signal processing slave modules set in a dual-redundancy manner, the signal processing standby module is a hot backup of the signal processing main module, and when the signal processing main module goes offline, the signal processing standby module goes online;

[0021] Such as Figure 2 shown, the signal processing main module includes a first switch, a first main processing node CPU and its own multiple first sub-processing nodes. The first main processing node CPU obtains the status information of its own multiple first sub-processing nodes through the first switch and generates a first information table;

[0022] Such asFigure 3 As shown, the signal processing standby module includes a second switch, a second main processing node CPU, and multiple second sub-processing nodes of its own. The second main processing node CPU obtains the status information of its multiple second sub-processing nodes through the second switch and generates a second information table;

[0023] As Figure 4 As shown, each signal processing slave module includes a third main processing node CPU, multiple third sub-processing nodes of its own, and a third switch. The third main processing node CPU obtains the status information of its multiple third sub-processing nodes through the third switch and generates a subnet information table;

[0024] The first main processing node CPU and the second main processing node CPU obtain the subnet information tables generated by all the signal processing slave modules, and summarize them with their own first information table or second information table into a backbone network information table respectively. The backbone network information table is read in real time by all signal processing slave modules, knows the working status of each node of other all signal processing slave modules and the signal processing master module, determines whether to perform data interaction. When the first main processing node CPU of the signal processing master module fails, the second main processing node CPU of the signal processing standby module manages the backbone network to solve the communication congestion problem caused by node offline.

[0025] Among them: The sub-processing node corresponds to a DSP (DSP is a digital signal processor, and the DSP is correspondingly connected to an external device) and an FPGA (an editable array processor, which is connected to an external device). The first main processing node CPU serves as the main node. The main processing node generates a subnet information table by obtaining the online status of each sub-processing node in real time. When the sub-processing node of a certain signal processing slave module is blocked, for example, DSP1 is blocked or offline (everyone knows that it is blocked and does not send messages), other nodes perform normal data interaction.

[0026] The design of the primary and standby signal processing modules solves the problem that each node of the rapidIO network cannot obtain the status of the peer node due to the failure of the signal processing master module, resulting in communication anomalies. The attempt to access the design solves the problem of network congestion caused by a large amount of communication data when the node goes offline. The rapidIO subnet design solves the problem that each point in the rapidIO subnet cannot communicate when both the primary and standby signal processing modules fail. Generally speaking, a rapidIO network management method for a signal processing system can be widely promoted and used as an implementation method.

[0027] As a specific implementation manner provided in this case, when the main processing node of the signal processing master module fails, the main processing node of the signal processing standby module manages the backbone network and attempts to access in a preset time interval to perform communication between nodes, solving the communication congestion problem caused by node offline.

[0028] As the specific implementation provided in this case, it further includes the following steps:

[0029] Step 1: Establish a subnet information table in each third main processing node CPU. The subnet information table records the network initialization status, subnet valid flag, online status of each node in the subnet, and physical link status of each node in the subnet;

[0030] Step 2: Each third main processing node CPU main node of each signal processing slave module obtains the status information of all its child nodes and generates a subnet information table, and the subnet information table is sent to the first main processing node CPU main section and the second main processing node CPU main section;

[0031] Step 3: After the first main processing node CPU main section and the second main processing node CPU main section obtain all the subnet information tables, they are respectively integrated with the corresponding first information table and second information table within a preset time period to form a backbone network information table, and the backbone network node information table is placed in the local shared memory area for all signal processing slave modules to access;

[0032] Step 4: At the beginning of each cycle, each third main processing node CPU of each signal processing slave module reads the backbone network information table from the first main processing node CPU to the local and distributes it to each of its child nodes.

[0033] As the specific implementation provided in this case, it further includes the following steps:

[0034] When the first main processing node CPU goes offline, the second main processing node CPU goes online. If the second main processing node CPU goes offline, the backbone network information table accessed by the signal processing slave module is marked as invalid, and the currently used rapidIO network enters the degraded working mode and no longer has network management capabilities; the subnet information table of each signal processing slave module is only used for communication between its own sub-processing nodes, or the third main processing node CPU adopts a strategy of attempting to access to re-establish a connection with the first main processing node CPU or the second main processing node CPU.

[0035] As the specific implementation provided in this case, in the currently used rapidIO network, the node resets the port connected to the peer node and reads the initialization completion flag of the peer node once. If it can be read, the access attempt is successful; otherwise, the processor generates data as an access exception and does not access the peer node in this cycle, solving the data congestion problem caused by continuously accessing the offline node.

[0036] As the specific implementation provided in this case, the first main processing node CPU or the second main processing node CPU maintains all the reported subnet information tables, including:

[0037] Detect the initialization completion flag. If the flag exists, it indicates the existence of the corresponding subnet information table, and it is detected only once. Then, detect the heartbeat signal. When there is no heartbeat, set the subnet information table to invalid. (The heartbeat signal is 1, 2, 3, 4... n, and each number represents the update status of all subnet information tables in different cycles). When there is no heartbeat, set the subnet information table to invalid.

[0038] As described above, it is only the specific implementation manner of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present disclosure should be covered within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A rapidIO network management method for a signal processing system, characterized in that, The signal processing main module, signal processing standby module and multiple signal processing slave modules are set in a dual-redundancy mode. The signal processing standby module is a hot standby for the signal processing main module. When the signal processing main module goes offline, the signal processing standby module goes online; The signal processing main module includes a first switch, a first main processing node CPU and multiple first sub-processing nodes. The first main processing node CPU obtains the status information of the multiple first sub-processing nodes through the first switch and generates a first information table; The signal processing standby module includes a second switch, a second main processing node CPU and multiple second sub-processing nodes. The second main processing node CPU obtains the status information of the multiple second sub-processing nodes through the second switch and generates a second information table; Each signal processing slave module includes a third switch, a third main processing node CPU and multiple third sub-processing nodes. The third main processing node CPU obtains the status information of the multiple third sub-processing nodes through the third switch and generates a subnet information table; The first main processing node CPU and the second main processing node CPU obtain the subnet information tables generated by all the signal processing slave modules, and respectively summarize them with their own first information table or second information table into a backbone network information table. The backbone network information table is read in real time by all the signal processing slave modules to know the working status of each node of all other signal processing slave modules and the signal processing main module, and determine whether to perform data interaction; When the first main processing node CPU of the signal processing main module fails, the second main processing node CPU of the signal processing standby module manages the backbone network to solve the communication congestion problem caused by node offline; A subnet information table is established in each third main processing node CPU. The subnet information table records the sub-network initialization status, subnet valid flag, online status of each node in the subnet, and physical link status of each node in the subnet; The third main processing node CPU of each signal processing slave module obtains the status information of all its own sub-nodes and generates a subnet information table, and the subnet information table is sent to the first main processing node CPU and the second main processing node CPU; After the first main processing node CPU and the second main processing node CPU obtain all the subnet information tables, they are respectively integrated with their corresponding first information table and second information table within a preset time period to form a backbone network information table, and the backbone network node information table is placed in the local shared memory area for all signal processing slave modules to access; At the beginning of each cycle, the third main processing node CPU of each signal processing slave module reads the backbone network information table from the first main processing node CPU to the local and distributes it to each of its own sub-nodes.

2. The method according to claim 1, characterized in that, When the main processing node of the signal processing main module fails, the main processing node of the signal processing standby module manages the backbone network and attempts to access in a preset time interval to perform communication between nodes to solve the communication congestion problem caused by node offline.

3. The method according to claim 1, wherein It also includes the following steps: When the first main processing node CPU goes offline, the second main processing node CPU goes online. If the second main processing node CPU goes offline, the backbone network information table accessed by the signal processing slave module is marked as invalid, and the currently used rapidIO network enters the degraded working mode and no longer has the backbone network management ability; the subnet information table of each signal processing slave module is only used for communication between its own sub-processing nodes, or the third main processing node CPU adopts a strategy of attempting to access to re-establish a connection with the first main processing node CPU or the second main processing node CPU.

4. The method according to claim 3, wherein In the currently used rapidIO network, the node resets the port connected to the peer node and reads the initialization completion flag of the peer node once. If it can be read, the access attempt is successful; otherwise, a data access exception is generated by the processor and the peer node is not accessed in this cycle, solving the data congestion problem caused by continuously accessing the offline node.

5. The method according to claim 4, wherein The first main processing node CPU or the second main processing node CPU maintains all the reported subnet information tables, including: Detect the initialization completion flag. The existence of the flag indicates the existence of the corresponding subnet information table, and it is detected only once. Then, the heartbeat signal is detected. When there is no heartbeat, the subnet information table is set to invalid.

Images