Distributed odd-even toughness networking method, electronic equipment and storage medium

Abstract

According to the distributed odd-even toughness networking method, the electronic device and the storage medium, the nodes in the cluster network are divided into the odd nodes and the even nodes, the odd nodes and the even nodes are bound with different time slots, simultaneous receiving and transmitting among multiple nodes are achieved, and the reliability of the distributed odd-even toughness networking method is improved. The method has the advantages that a spread spectrum system is used, the method is suitable for large-scale transmission distance, complexity is low, toughness is high, and the method can particularly meet the characteristic requirements of high-orbit cluster networking.

Description

Technical Field

[0001] The present invention mainly relates to a cluster networking method, and in particular to a distributed parity resilience networking method, an electronic device and a storage medium. Background Art

[0002] High-orbit cluster networking has the characteristics of large transmission distance between nodes, low dynamics, high networking resilience requirements, the need to use a spread spectrum system, and fewer satellite node resources. Existing networking solutions cannot meet the characteristics of this type of network well, and often cannot take advantage of the spread spectrum system's anti-co-channel interference advantages. For example, the common CSMA / CA networking protocol (Carrier Sense Multiple Access) will greatly increase the probability of node collisions in CSMA / CA due to the high transmission delay caused by the large-scale transmission distance, resulting in poor network scalability. Summary of the invention

[0003] Based on the above problems, in a first aspect, the present application provides a distributed parity resilience networking method, which is applied to each node in a cluster network, and the method includes:

[0004] When the topology update of the cluster network is triggered, the current gender and its corresponding opposite-sex neighbor node are determined, where the current gender is either an odd node or an even node, and the odd node and the even node are bound to different time slots;

[0005] The comprehensive neighborhood is determined by the opposite-sex neighbor nodes, and the gender ratio within the comprehensive neighborhood is counted. The comprehensive neighborhood refers to the neighborhood centered on the current node and its corresponding opposite-sex neighbor nodes, and the gender ratio within the domain refers to the ratio between the number of nodes with the same gender as the current node and the total number of nodes in the comprehensive neighborhood.

[0006] Gender switching is performed according to the gender ratio in the domain and the preset allocation interval, and the link relationship in the comprehensive neighborhood is updated according to the target gender after switching.

[0007] In some embodiments, before determining the current gender and its corresponding opposite-sex neighbor node, the method further includes:

[0008] Determine the in-degree and out-degree in the current neighbor table, where the in-degree represents the number of neighbor nodes connected to the current node, and the out-degree represents the number of neighbor nodes connected to the current node;

[0009] Determine whether the in-degree is greater than or equal to the preset in-degree threshold, and determine whether the out-degree is greater than or equal to the preset out-degree threshold; wherein,

[0010] If the in-degree is less than the in-degree threshold, or the out-degree is less than the out-degree threshold, the current gender is switched to its corresponding opposite gender; otherwise, no processing is done.

[0011] In some embodiments, gender switching is performed according to the gender ratio in the domain and the preset allocation interval, including:

[0012] Start the judgment timer;

[0013] Determine whether the gender ratio in the domain belongs to the distribution interval; where:

[0014] If the gender ratio in the domain is less than the minimum allocation threshold of the allocation interval, the current gender is switched to the corresponding opposite gender;

[0015] If the gender ratio in the domain is greater than the minimum allocation threshold, and the gender ratio in the domain is less than the maximum allocation threshold of the allocation interval, the judgment timer is increased by a preset backoff value, and all nodes in the comprehensive domain are notified, so that all nodes in the comprehensive domain increase the backoff value on their own judgment timers;

[0016] If the gender ratio in the domain is greater than the maximum allocation threshold, the neighboring nodes of the opposite sex are notified to switch their gender to the corresponding opposite sex.

[0017] In some embodiments, the allocation interval is obtained by:

[0018] Convert the topological structure of the cluster network into a directed graph, which includes all nodes in the cluster network;

[0019] The network redundancy of the directed graph is used as the evaluation index, and the correlation between the evaluation index and the allocation interval is defined, where the network redundancy represents the weighted sum of the average number of links between all node pairs in the directed graph;

[0020] The association relationship is globally solved through simulation correction until the evaluation index reaches the optimal level and the corresponding allocation interval is obtained.

[0021] In some embodiments, the expression of network redundancy is as follows:

[0022]

[0023] Where NR is the network redundancy; i and j are the two subscripts of a node pair in a directed graph; N is the total number of nodes, trace() is an operation function used to calculate the trace of the matrix; A is the adjacency matrix corresponding to the topological structure, R is the intermediate matrix, and the expression of R is represents the number of redundant links between all node pairs, k is the path length, and K is the maximum path length.

[0024] In some embodiments, the evaluation index further includes network efficiency, wherein the network efficiency represents the average value of the inverse of the shortest path lengths between all node pairs; and the median value of the allocation interval is 0.5.

[0025] In some embodiments, after updating the link relationship in the comprehensive neighborhood according to the switched target gender, the method further includes:

[0026] The path arrival status of the current node is determined through the consistency protocol between nodes, where the expression of the path arrival status is x i [N], i is the subscript of the current node, N is the number of information exchanges with adjacent nodes, which is consistent with the total number of nodes in the cluster network;

[0027] According to the path arrival status, the cluster network is strongly connected to detect the strong connectivity, and the strong connectivity detection status of the cluster network is obtained. The expression of the strong connectivity status is f i [N], i is the subscript of the current node, N is the number of information exchanges with adjacent nodes, which is consistent with the total number of nodes in the cluster network;

[0028] When the strong connectivity detection status indicates strong connectivity, the gender switching of the current node is terminated to keep the cluster network stable.

[0029] In some embodiments, the path arrival state is obtained by performing N iterations of updating of the path arrival initial value, and the expression of each update operation is as follows:

[0030] The expression of the path to the initial value as follows:

[0031]

[0032] The strong connectivity detection state is obtained by updating the strong connectivity detection initial value N times through iterations. The expression of each update operation is as follows:

[0033] The expression of the initial value of strong connectivity detection is as follows:

[0034]

[0035] Where j is the subscript of the adjacent node, and t is the ordinal number corresponding to the tth information interaction.

[0036] In a second aspect, the present application provides an electronic device, including a memory and a processor, wherein instructions are stored in the memory, and when the instructions are called by the processor, the processor executes any method as described in the first aspect above.

[0037] In a third aspect, the present application provides a storage medium comprising computer program instructions, wherein the computer program instructions are used to enable a computer to execute any method of the first aspect.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] The purpose of the present invention is to provide a distributed odd-even resilient networking method, electronic device and storage medium, which divides the nodes in the cluster network into odd nodes and even nodes, and binds the odd nodes and even nodes to different time slots, so as to realize simultaneous transmission and reception between multiple nodes. It has the characteristics of using a spread spectrum system, being suitable for large-scale transmission distances, having low complexity and strong resilience, and is particularly adaptable to the characteristic requirements of high-orbit cluster networking. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings are included to provide a further understanding of the present application. They are included and constitute a part of the present application. The accompanying drawings illustrate embodiments of the present application and together with the present specification serve to explain the principles of the present application. In the accompanying drawings:

[0041] Figure 1 This is a flow chart of a distributed parity resilience networking method exemplarily provided by the present application;

[0042] Figure 2 is a schematic diagram of a topological structure exemplarily provided in this application;

[0043] Figure 3 It is a logical diagram of an even-odd backoff switching strategy exemplarily provided in the present application;

[0044] Figure 4a , Figure 4b This is a relationship diagram between gender ratio and network redundancy provided by the present application as an example;

[0045] Figure 5a , Figure 5b This is a graph showing the relationship between gender ratio and network efficiency provided as an example in this application;

[0046] Figure 6 This is a throughput comparison chart of a simulation verification provided by the present application as an example;

[0047] Figure 7a , Figure 7b This is a rendering of a distributed parity backoff switching strategy exemplarily provided in this application;

[0048] Figure 8 This is one of the toughness capability comparison diagrams provided by the present application for illustrative purposes by simulation verification;

[0049] Fig. 9 This is the second comparison chart of toughness capability verified by simulation provided by the present application as an example;

[0050] Fig.10 This is a schematic diagram of an electronic device provided as an example in this application. DETAILED DESCRIPTION

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0052] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0053] In a first aspect of this embodiment, a distributed parity resilience networking method is provided. Figure 1 The method can be applied to the field of cluster networking technology. The cluster may refer to an on-orbit service (OOS) cluster composed of multiple high-orbit satellites, and any high-orbit satellite in the cluster is used as a node to form a communication network. The steps of a distributed parity resilience networking method according to this embodiment can be performed on any node in this cluster network. Figure 1 , taking any node as the execution subject, the steps of a distributed parity resilience networking method provided by the present application include:

[0054] S101, when a topology update of a cluster network is triggered, a current gender and its corresponding opposite-sex neighbor node are determined, wherein the current gender is any one of an odd node and an even node, and the odd node and the even node are bound to different time slots.

[0055] S102, determining a comprehensive neighborhood through opposite-sex neighbor nodes, and counting the gender ratio within the comprehensive neighborhood, wherein the comprehensive neighborhood refers to a neighborhood centered on the current node and its corresponding opposite-sex neighbor node, and the gender ratio within the domain refers to the ratio between the number of nodes with the same gender as the current node and the total number of nodes in the comprehensive neighborhood.

[0056] S103, performing gender switching according to the gender ratio in the domain and the preset allocation interval, and updating the link relationship in the comprehensive neighborhood according to the switched target gender.

[0057] refer to Figure 2, which is a schematic diagram of a topological structure provided by this embodiment. In the cluster network, the gender of the nodes is divided into odd nodes and even nodes, and the communication cycle of the entire network is correspondingly divided into only two odd and even time slots. The odd nodes are bound to the odd time slots, and the even nodes are bound to the even time slots. In the odd time slots, all odd nodes broadcast, and each even node receives messages from nearby odd nodes; while in the even time slots, all even nodes broadcast, and each odd node receives messages from nearby even nodes.

[0058] One aspect of the above steps is to achieve global uniformity through local uniformity of the gender of nodes in the entire network. This application refers to this as a distributed parity backoff switching strategy. In actual networking, each node in the cluster network can obtain its own neighbor information (usually two hops) through the interaction of periodic HELLO packets, so as to determine the current gender and its corresponding opposite-sex neighbor node when triggering the topology update of the cluster network. The topology update can be implemented by a timer maintained on each node, for example, setting the timer to a set range [h min ,h max ] and starts counting down from the random value, where h min The minimum value for setting the timing, h max To set the maximum value of the timing, when the timer value is 0, the node confirms that the topology structure needs to be updated. Random values ​​can avoid the problem of synchronization degeneration caused by multiple nodes in the network performing odd-even switching at the same time. The opposite-sex neighbor node can be any node among all the bidirectional neighbor nodes of the current node.

[0059] In some embodiments, before determining the current gender and its corresponding opposite-sex neighbor nodes, it also includes: determining the in-degree and out-degree in the current neighbor table, wherein the in-degree represents the number of neighbor nodes connected to the current node, and the out-degree represents the number of neighbor nodes connected to the current node; judging whether the in-degree is greater than or equal to a preset in-degree threshold, and judging whether the out-degree is greater than or equal to a preset out-degree threshold; wherein, if the in-degree is less than the in-degree threshold, or the out-degree is less than the out-degree threshold, the current gender is switched to its corresponding opposite sex; otherwise, no processing is performed.

[0060] One aspect of the above steps is to enable the node to determine whether the gender distribution in the current field meets the allocation requirements before confirming the comprehensive neighborhood. If not, the node can directly switch gender.

[0061] In some embodiments, gender switching is performed according to the gender ratio in the domain and a preset allocation interval, including: starting a judgment timer; judging whether the gender ratio in the domain belongs to the allocation interval; wherein: if the gender ratio in the domain is less than the minimum allocation threshold of the allocation interval, the current gender is switched to the corresponding opposite sex; if the gender ratio in the domain is greater than the minimum allocation threshold, and the gender ratio in the domain is less than the maximum allocation threshold of the allocation interval, the judgment timer is increased by a preset backoff value, and all nodes in the comprehensive domain are notified, prompting all nodes in the comprehensive domain to also increase the backoff value on their own judgment timers; if the gender ratio in the domain is greater than the maximum allocation threshold, the opposite-sex neighboring nodes are notified to switch their gender to the corresponding opposite sex.

[0062] The determination timer may be another timer maintained by the node, or the above timer may be reset, and there is no limitation on this.

[0063] refer to Figure 3 , which is a logical diagram of an odd-even backoff switching strategy exemplarily provided in this embodiment. Taking the current node as an odd node i as an example, the node first checks its own in-degree and out-degree, that is, checks the number of its in-neighbors and out-neighbors. If one of them is less than the threshold z, it proves that the parity distribution in the area where the node is located is uneven, and the node directly switches gender. If both the in-degree and the out-degree are greater than or equal to the threshold z, node i randomly selects a node j (node ​​j must be an even node) as a neighbor node of the opposite sex from the two-way neighbor stage, and sends a neighborhood information request to j. After receiving the neighborhood information request, j sends the neighborhood information associated with j to i. i integrates its own and j's neighborhood information to determine a comprehensive neighborhood centered on i and j, and counts the number of odd nodes in the comprehensive neighborhood to obtain the ratio between this number and the total number of nodes, that is, the gender ratio q in the domain. The minimum allocation threshold and the maximum allocation threshold of the allocation interval are q respectively. min ,q max When q min When q>q, the number of odd nodes in the comprehensive neighborhood is small. At this time, node i notifies node j to switch gender and resets the judgment timer. max When q=[q min ,q max ], the parity distribution in the comprehensive neighborhood meets the allocation requirements, and node i increases its judgment timer by the backoff value h bk , and notify all nodes in the integrated neighborhood, prompting all nodes in the integrated area to increase the value of the judgment timer by the backoff value h bk .

[0064] The purpose of increasing the backoff value is to ensure that once a certain area reaches a uniform parity distribution, it remains stable for a period of time to prevent unnecessary network overhead caused by frequent parity switching.​

[0065] In some embodiments, the allocation interval is obtained in the following manner: the topological structure of the cluster network is converted into a directed graph, which includes all nodes in the cluster network; the network redundancy of the directed graph is used as an evaluation index, and the association relationship between the evaluation index and the allocation interval is defined, wherein the network redundancy represents the weighted sum of the average number of links between all pairs of nodes in the directed graph; the association relationship is globally solved through simulation correction until the evaluation index reaches the optimal level, thereby obtaining the corresponding allocation interval.

[0066] Network redundancy can monotonically and sensitively reflect the redundancy of edges in the network, thereby effectively evaluating the connectivity of directed networks. The only prerequisite is that the topological structure of the network itself needs to be converted into a directed graph. Therefore, in addition to evaluating the high-rail cluster network as embodied in this embodiment, network redundancy itself can also be evaluated for other network designs that can be converted into directed graphs, such as drone networking, etc., and this application does not limit this.

[0067] In some embodiments, the expression of network redundancy is shown in the following formula (1):

[0068]

[0069] is the total number of nodes, trace() is an operation function used to calculate the trace of the matrix; A is the adjacency matrix corresponding to the topological structure, R is the intermediate matrix, and the expression of R is represents the number of redundant links between all node pairs, k is the path length, and K is the maximum path length.

[0070] This application does not elaborate on the derivation process of network redundancy. In essence, network redundancy is intended to count the number of links between pairs of nodes in the network, and takes into account the impact of short self-loops, so it can correctly reflect the connectivity of the directed network, and thus has a strong advantage in meeting the requirements of network resilience. Using network redundancy as an evaluation indicator, we can define its correlation with the allocation interval, and solve it through simulation correction, refer to Figure 4a In the simulation scenario with a total of 50 nodes, the horizontal axis represents the gender ratio in the network, and the vertical axis represents the network redundancy. From top to bottom, the five curves correspond to 9, 8, 7, 6, and 5 receiving channels respectively. Figure 4b , which corresponds to a simulation scenario with a total number of nodes of 100. Figure 4b The horizontal axis in the figure represents the gender ratio in the network, and the vertical axis represents the network redundancy. From top to bottom, the five curves correspond to 9, 8, 7, 6, and 5 receiving channels respectively. Figure 4a and Figure 4bThe results show that when the gender ratio in the network is around 50%, the evaluation index reaches the optimal value, so the effect is best when the odd and even distribution in the network is even. At the same time, the results also show that as the total number of nodes increases, the distribution interval corresponding to the optimal evaluation index is also increasing, which means that the more the total number of nodes, the more tolerant the network redundancy is to the gender ratio.

[0071] In some embodiments, the evaluation index also includes network efficiency, where network efficiency represents the average value of the reciprocal of the shortest path length between all node pairs. The correlation between gender ratio and network efficiency is as follows: Figure 5a , Figure 5b As shown, Figure 5a , Figure 5b These correspond to simulation scenarios with a total number of nodes of 50 and 100, respectively. Figure 5a , Figure 5b The horizontal axis in the figure represents the gender ratio in the network, and the vertical axis represents the network efficiency. Figure 5a and Figure 5b ,The results show that when the gender ratio in the network is around 50%, the evaluation index reaches the best ,trend, which is basically consistent with the trend of network redundancy.

[0072] Therefore, in actual networking, the median value of the allocation interval can be set at 0.5. In other words, the number and spatial distribution of odd nodes and even nodes in the network can be as uniform as possible, so as to better meet the performance requirements of the network.

[0073] In this embodiment, the median value of the allocation interval is 0.5. However, if the allocation interval is small, it may cause parity switching oscillation problems in some areas of the network, and if the allocation interval is large, it may cause a certain degree of network performance loss. Therefore, the following distributed strong connectivity judgment strategy is provided to solve this problem, including: determining the path arrival state of the current node through the consistency protocol between each node, where the expression of the path arrival state is x i [N], i is the subscript of the current node, N is the number of information interactions with adjacent nodes, which is consistent with the total number of nodes in the cluster network; the cluster network is strongly connected according to the path arrival status to obtain the strong connectivity detection status of the cluster network, where the expression of the strong connectivity status is f i [N], i is the subscript of the current node, N is the number of information interactions with adjacent nodes, which is consistent with the total number of nodes in the cluster network; when the strong connectivity detection status indicates strong connectivity, the gender switching of the current node is terminated to keep the cluster network stable.

[0074] In the distributed strong connectivity judgment strategy, each node in the cluster network maintains the state x i [t] and f i [t],x i[t] is used to verify whether node i can be reached from any other node, f i [t] is used to verify whether the network is strongly connected, and t is the ordinal number corresponding to the tth information interaction. Therefore, the path arrival state is obtained by iterating N times the path arrival initial value. The node updates x according to the consistency protocol. i The jth row x in [t] i,j [t], the expression of each update operation is shown in the following formula (2):

[0075]

[0076] The expression of the path reaching the initial value is shown in the following formula (3):

[0077]

[0078] Each node verifies locally whether x is satisfied. i [N] = 1 N , and update the state f according to the consistency protocol i [t]. Therefore, the strong connectivity detection state is obtained by updating the initial value of the strong connectivity detection N times through iterations. The expression of each update operation is as follows:

[0079]

[0080] The expression of the initial value of strong connectivity detection is shown in the following formula (5):

[0081]

[0082] Where j is the subscript of the adjacent node.

[0083] By adding a distributed strong connectivity judgment algorithm to the parity switching strategy, a strong connectivity judgment is performed every period of time. If the network is a strongly connected graph, the parity switching strategy is stopped to keep the network stable.

[0084] Since distributed parity resilient networking can use the parity switching strategy to dynamically change the gender of nodes and thus change the topological connection relationship of the network, it also has certain advantages in network resilience. Regarding the resilience evaluation model, the more comprehensive Tran model can be used, which will not be repeated here.

[0085] The following provides a simulation verification of the distributed parity resilience networking method in an actual scenario. The simulation parameters are shown in Table 1:

[0086] Table 1

[0087]

[0088]

[0089] The simulation verification compares the throughput of three schemes: TDMA (time division multiple access), CSMA / CA and distributed parity resilience networking as the number of network nodes and network load change. Figure 6 , from top to bottom they correspond to distributed parity resilient networking, CSMA / CA and TDMA. Figure 6 The x, y, and z axes represent the load, number of nodes, and throughput (in log10 form), respectively. Figure 6 It can be seen that under the parity-resilient networking scheme, the throughput continues to increase with the increase in the number of nodes and network load. This is because it fully utilizes the advantages of the spread spectrum system. Nearly half of the nodes in the network are sending data packets in each time slot, and the other half of the nodes can also receive data packets from 5 nodes at the same time. The throughput of this scheme will increase with the increase in the number of nodes and is not limited by the number of nodes. Figure 7a , Figure 7b The effect of the distributed parity backoff switching strategy is shown, where the x-axis and y-axis represent time, network redundancy or network efficiency respectively. Figure 7a The network redundancy change curve shows that in the networking process using the distributed parity backoff switching strategy, the network redundancy of the network is significantly improved in a short period of time, and then remains stable. Figure 7b The network efficiency change curves shown also show a basically consistent trend, reflecting that the network topology reaches stability in a short period of time. Figure 8 The resilience capabilities of the parity resilience networking solutions are compared from the perspective of throughput, with the x and y axes representing time and resilience values, respectively. From top to bottom, they correspond to the three situations of parity non-switching, parity random switching, and parity backoff switching. Figure 8 It can be seen that the parity backoff switching strategy is higher than the other two in terms of average throughput and average resilience, and has moderate volatility. This is because the backoff switching takes into account the uniformity of the number and space of parity nodes and uses the backoff mechanism to ensure the stability of the parity switching; the average throughput of the random switching strategy is lower and the volatility is larger. This is because although the random switching can guarantee the uniformity of the number of parity nodes in the network in terms of probability, it cannot guarantee the uniformity of the spatial distribution, and the absence of the backoff mechanism makes the frequency of parity switching higher, and the network topology is consistently in an unstable state; the non-switching strategy first uses the backoff strategy to obtain a network with uniform parity nodes and then stops switching, so the initial performance of the network is similar to the backoff switching, and the parity nodes of the network after the random node attack are still evenly distributed, so even if the parity switching is not performed, its performance will not drop much, and the network is stable, but when a single-sex node attack is carried out on the network, the network performance will drop significantly or even collapse. Fig. 9The resilience capability of the parity-resilient networking solution is demonstrated from the perspective of network redundancy, and its overall trend is consistent with the throughput. Since the network redundancy value can be negative, and the existing resilience evaluation system (reference: "A framework for the quantitative assessment of performance-based system resilience") is only applicable to positive numbers, the network redundancy data is shifted upward by 5 when calculating the resilience value. From top to bottom, they correspond to the three situations of no parity switching, random parity switching, and parity backoff switching. Therefore, the results show that the average redundancy and average resilience of the backoff switching strategy are higher than the other two, proving that the networking solution proposed in this application has great advantages in throughput and resilience.

[0090] Further, if Fig.10 An exemplary embodiment of the present application further provides an electronic device, including a memory 1001 and a processor 1002, wherein the memory 1001 stores instructions, and when the instructions are executed by the processor 1002, the processor 1002 executes any method of the first aspect.

[0091] It should be understood that the processor mentioned in the embodiments of the present application may be a CPU, or other general-purpose processors, DSPs, ASICs, FPGAs or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0092] It should also be understood that the memory mentioned in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory, an erasable programmable read-only memory, an electrically erasable programmable read-only memory, or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static random access memory, dynamic random access memory, synchronous dynamic random access memory, double data rate synchronous dynamic random access memory, enhanced synchronous dynamic random access memory, synchronously connected dynamic random access memory, and direct memory bus random access memory.

[0093] The present application also provides a storage medium storing computer program instructions, which, when executed by a processor of a computer, enables the computer to execute the steps of any of the methods mentioned above.

[0094] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or any combination of the above mediums.

[0095] The basic concepts have been described above. Obviously, for those skilled in the art, the above invention disclosure is only used as an example and does not constitute a limitation of the present application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements and amendments to the present application. Such modifications, improvements and amendments are suggested in the present application, so such modifications, improvements and amendments still belong to the spirit and scope of the exemplary embodiments of the present application.

[0096] At the same time, the present application uses specific words to describe the embodiments of the present application. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of the present application. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present application can be appropriately combined.

[0097] Some aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". The processor may be one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DAPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. In addition, various aspects of the present application may be expressed as computer products located in one or more computer-readable media, which include computer-readable program codes. For example, computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, tapes ...), optical disks (e.g., compact disks CDs, digital versatile disks DVDs ...), smart cards, and flash memory devices (e.g., cards, sticks, key drives ...).

[0098] A computer-readable medium may include a propagated data signal containing computer program code, such as in baseband or as part of a carrier wave. The propagated signal may have a variety of manifestations, including electromagnetic, optical, etc., or a suitable combination. A computer-readable medium may be any computer-readable medium other than a computer-readable storage medium, which may be connected to an instruction execution system, device or apparatus to communicate, propagate or transmit a program for use. The program code on the computer-readable medium may be propagated via any suitable medium, including radio, cable, fiber optic cable, radio frequency signal, or similar medium, or any combination of the above mediums.

[0099] Similarly, it should be noted that in order to simplify the description of the disclosure of this application and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this application, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the object of this application are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0100] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of the present application are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0101] Although the present application has been described with reference to the current specific embodiments, ordinary technicians in this technical field should recognize that the above embodiments are only used to illustrate the present application, and various equivalent changes or substitutions may be made without departing from the spirit of the present application. Therefore, as long as the changes and modifications to the above embodiments are within the essential spirit of the present application, they will fall within the scope of the claims of the present application.

Claims

1. A distributed parity resilience networking method, characterized in that: Applied to each node in a cluster network, the method comprises: When the topology structure update of the cluster network is triggered, the current gender and its corresponding opposite-sex neighbor node are determined, wherein the current gender is any one of an odd node and an even node, and the odd node and the even node are bound to different time slots; Determine a comprehensive neighborhood through the opposite-sex neighbor nodes, and count the gender ratio within the comprehensive neighborhood, wherein the comprehensive neighborhood refers to a neighborhood centered on the current node and its corresponding opposite-sex neighbor nodes, and the gender ratio within the domain refers to the ratio between the number of nodes in the comprehensive neighborhood that are consistent with the current gender and the total number of nodes; Gender switching is performed according to the gender ratio in the domain and the preset allocation interval, and the link relationship in the comprehensive neighborhood is updated according to the target gender after the switching.

2. The distributed parity resilient networking method according to claim 1, characterized in that: Before determining the current gender and its corresponding opposite-sex neighbor node, the method further includes: Determine the in-degree and out-degree in the current neighbor table, wherein the in-degree represents the number of neighbor nodes connected to the current node, and the out-degree represents the number of neighbor nodes connected to the current node; Determine whether the in-degree is greater than or equal to a preset in-degree threshold, and determine whether the out-degree is greater than or equal to a preset out-degree threshold; wherein, If the in-degree is less than the in-degree threshold, or the out-degree is less than the out-degree threshold, the current gender is switched to its corresponding opposite gender; otherwise, no processing is performed.

3. The distributed parity resilience networking method according to claim 2, characterized in that: The performing gender switching according to the gender ratio in the domain and the preset allocation interval includes: Start the judgment timer; Determine whether the gender ratio in the domain belongs to the allocation interval; wherein: If the gender ratio in the domain is less than the minimum allocation threshold of the allocation interval, switching the current gender to the corresponding opposite gender; If the gender ratio in the domain is greater than the minimum allocation threshold, and the gender ratio in the domain is less than the maximum allocation threshold of the allocation interval, the judgment timer is increased by a preset backoff value, and all nodes in the comprehensive domain are notified, so that all nodes in the comprehensive domain increase the backoff value on their own judgment timers; If the gender ratio in the domain is greater than the maximum allocation threshold, the opposite-sex neighbor node is notified to switch its gender to the corresponding opposite sex.

4. The distributed parity resilience networking method according to any one of claims 1 to 3, characterized in that: The allocation interval is obtained in the following manner: Converting the topological structure of the cluster network into a directed graph, wherein the directed graph includes all nodes in the cluster network; Using the network redundancy of the directed graph as an evaluation index, defining an association relationship between the evaluation index and the allocation interval, wherein the network redundancy represents a weighted sum of the average number of links between all pairs of nodes in the directed graph; The association relationship is globally solved through simulation correction until the evaluation index reaches the optimum, thereby obtaining the corresponding allocation interval.

5. The distributed parity resilience networking method according to claim 4, characterized in that: The expression of the network redundancy is as follows: Wherein, NR is the network redundancy; i and j are the two subscripts of a node pair in the directed graph; N is the total number of nodes, trace() is an operation function used to calculate the trace of the matrix; A is the adjacency matrix corresponding to the topological structure, R is the intermediate matrix, and the expression of R is represents the number of redundant links between all node pairs, k is the path length, and K is the maximum path length.

6. The distributed parity resilience networking method according to claim 4, characterized in that: The evaluation index also includes network efficiency, wherein the network efficiency represents the average value of the reciprocal of the shortest path length between all node pairs; and the median value of the allocation interval is 0.

5.

7. The distributed parity resilience networking method according to claim 4, characterized in that: After updating the link relationship in the comprehensive neighborhood according to the switched target gender, the method further includes: The path arrival state of the current node is determined by the consistency protocol between the nodes, wherein the expression of the path arrival state is x i [N], i is the subscript of the current node, N is the number of information interactions with adjacent nodes, which is consistent with the total number of nodes in the cluster network; According to the path arrival state, the cluster network is subjected to strong connectivity detection to obtain the strong connectivity detection state of the cluster network, wherein the expression of the strong connectivity state is f i [N], i is the subscript of the current node, N is the number of information interactions with adjacent nodes, which is consistent with the total number of nodes in the cluster network; When the strong connectivity detection state indicates strong connectivity, the gender switching of the current node is stopped to keep the cluster network stable.

8. The distributed parity resilience networking method according to claim 6, characterized in that: The path arrival state is obtained by performing N iterative updates on the path arrival initial value, and the expression of each update operation is as follows: The expression of the path reaching the initial value is as follows: The strong connectivity detection state is obtained by updating the strong connectivity detection initial value N times through iterations. The expression of each update operation is as follows: The expression of the initial value of the strong connectivity detection is as follows: Where j is the subscript of the adjacent node, and t is the ordinal number corresponding to the tth information interaction.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores instructions, and when the instructions are called by the processor, the processor executes the method according to any one of claims 1 to 8.

10. A storage medium, characterized in that: The method comprises computer program instructions, wherein the computer program instructions are used to cause a computer to execute the method according to any one of claims 1 to 8.

Images