Method and system for large-scale federated simulation interest management based on von neumann diagram

Abstract

The application relates to a large-scale joint simulation interest management method and system based on a von neumann diagram. The method comprises the following steps: setting a first accessed matching node as a network management node, controlling new node access by the network management node, generating a von neumann diagram based on existing node distribution, dividing simulation data management areas of all nodes, and being responsible for corresponding area data transmission. A client sends an interested area data subscription request to any matching node, and after receiving the request, the node sets nodes with overlapping management areas and interested areas as main matching nodes. The client publishes simulation data to the main matching node, the main matching node notifies relevant main matching nodes according to the subscription record, and then sends the simulation data to the subscription client. The method can reduce redundant simulation data in network transmission, optimize transmission interaction relationship, improve communication efficiency of large-scale distributed simulation, and guarantee smooth operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of simulation technology, in particular to a large-scale joint simulation interest management method and system based on a von Neumann diagram. BACKGROUND

[0002] Distributed simulation technology can integrate simulation resources distributed in different regions, build a simulation system sharing unified time, space and virtual scene environment, and has irreplaceable application value in multiple key fields. Specifically, it can be applied to simulation deduction to support decision analysis, personnel training to improve operational ability, new equipment research and development to shorten the research and development cycle, and equipment system performance evaluation to optimize system performance, etc. It provides important technical support for the development of related fields.

[0003] With the continuous improvement of industry demand for simulation accuracy and scale, the scale of distributed simulation system is continuously expanding, and the number of simulation nodes (clients) participating in it increases significantly, which directly leads to exponential growth of simulation data based on network transmission. Under the realistic conditions of limited network bandwidth, processing capacity and other resources, how to ensure the smoothness and real-time performance of the entire distributed simulation system data transmission, and avoid affecting the simulation effect due to data transmission problems, has become a core technical problem that needs to be solved urgently, which restricts the further development of distributed simulation technology.

[0004] Practice shows that reducing the transmission and operation of redundant data in the network is the key path to improve the overall performance of the distributed simulation system in large-scale application scenarios, and is also the key direction of the development of this field of technology optimization. Therefore, it is of great significance to break through the existing technical bottlenecks in this direction.

[0005] However, the current large-scale distributed simulation system still has significant technical defects in data transmission and processing: on the one hand, the traditional C / S architecture is limited by the centralized design and cannot adapt to the demand for dynamic expansion of nodes, and it is also difficult to carry the efficient transmission of massive simulation data, and has been unable to meet the application requirements of large-scale systems; on the other hand, the publish / subscribe mode commonly used in distributed simulation systems can realize data transmission and reception, but the traditional scheme has many shortcomings: first, the rule grid unit division method is used, and when the user subscribes to the cells intersecting the area of interest (AOI) to obtain information, it is easy to receive redundant data beyond the actual needs. If the grid is reduced to improve the subscription accuracy, the subscription maintenance cost will be greatly increased, causing waste of CPU processing resources; second, the static partition strategy is generally used, and when the user node position changes, the system is difficult to update the topology in real time. In the scene of a large number of nodes suddenly flowing into the local area, the system overload problem may also occur; third, the publish / subscribe mechanism is based on the topic (Topic) design, and for spatial attribute information, how to scientifically divide the space and effectively associate the space with the transmission channel has not formed a perfect solution, and the static partition characteristics of the existing interest management method cannot update the related configuration in time after the simulation node position changes. In the scene where the number of simulation entities is large or the motion trajectory of the entities is unpredictable, the practicability of the scheme is greatly reduced. SUMMARY

[0006] Therefore, it is necessary to provide a large-scale joint simulation interest management method and system based on von Neumann diagram, which can reduce the transmission of redundant simulation data and effectively improve the communication efficiency.

[0007] A large-scale joint simulation interest management method based on von Neumann diagram, the method is implemented in a large-scale joint simulation system including matching nodes, network management nodes and client nodes, the method comprises:

[0008] The first matching node accessing the system is configured as a network management node, which determines whether the matching node to be accessed can join the system, and generates a von Neumann diagram based on the distribution information of all matching nodes that have joined the system, and divides the simulation data management area of each matching node according to the von Neumann diagram, and each matching node is responsible for data sending and receiving in the corresponding simulation data management area;

[0009] The client node sends a subscription request for the area of interest data to any matching node, and the matching node receiving the subscription request broadcasts the subscription request, and sets the matching node whose simulation data management area overlaps with the area of interest as the primary matching node of the corresponding client node;

[0010] The client node publishes simulation data to the corresponding master matching node, and the master matching node informs the master matching nodes of the relevant subscription client nodes according to the subscription record, and the simulation data is sent to all relevant subscription client nodes by the master matching node.

[0011] In one embodiment, the simulation data management area of each matching node is divided according to the von Neumann diagram, including:

[0012] According to the number of all matching nodes in the system, the virtual space is divided into a corresponding number of initial simulation data management areas, so that each matching node is the nearest node to all nodes in its area according to the von Neumann diagram.

[0013] According to the capacity limit of each matching node, the area shape and size of each initial simulation data management area are optimized to obtain an optimized simulation data management area that adapts to the node load.

[0014] Based on the optimized simulation data management area, a tree structure covering the neighbors of the region of interest of all matching nodes is constructed according to the extended tree rule.

[0015] According to the optimized simulation data management area and the tree structure, data transmission is realized.

[0016] In one embodiment, based on the optimized simulation data management area, a tree structure covering the neighbors of the region of interest of all matching nodes is constructed according to the extended tree rule, including:

[0017] For each matching node, among all neighboring matching nodes of the matching node, the parent node of the matching node and the target node are excluded, and then the nodes overlapping with the region of interest of the target node are selected.

[0018] The node with the smallest distance to the target node among the neighboring matching nodes of the selected node is calculated, and if the node with the smallest distance is the current matching node, the selected node is determined as the child node of the current matching node.

[0019] Repeat the above process until all neighbors of the region of interest of all matching nodes are included in the tree structure, and ensure that there is no case where two nodes in the extended tree share the same parent node.

[0020] In one embodiment, when the matching node whose simulation data management area overlaps with the region of interest is set as the master matching node of the corresponding client node:

[0021] The matching node corresponding to at least one simulation data management area covered by the region of interest is set as the master matching node.

[0022] When the region of interest covers multiple simulation data management regions, the subscription request is pushed to the adjacent matching node backup subscription information by the main matching node.

[0023] In one embodiment, after the client node publishes simulation data to the corresponding main matching node;

[0024] If the region of interest associated with the published simulation data overlaps with the simulation data management region responsible by the adjacent matching node of the main matching node of the relevant subscription client node, the simulation data is pushed to all adjacent matching nodes whose simulation data management regions overlap with the region of interest associated with the published data, and the simulation data is published to the corresponding client node by the adjacent matching node.

[0025] In one embodiment, in the process of large-scale joint simulation interest management, if a matching node does not receive any subscription request within a preset time period, the matching node automatically leaves the system;

[0026] Meanwhile, the network management node adjusts the simulation data management region responsible by the adjacent matching node of the matching node leaving the system.

[0027] In one embodiment, when the client node sends a subscription request for the region of interest data to any matching node, if there is no matching node in the system, the client node directly sends a subscription request to the gateway node.

[0028] The application also provides a large-scale joint simulation interest management system based on von Neumann diagram, which comprises matching nodes, network management nodes and client nodes, and in the system:

[0029] The first matching node accessing the system is configured as a network management node, which determines whether the matching node to be accessed can join the system, generates a von Neumann diagram based on the distribution information of all matching nodes that have joined the system, and divides the simulation data management region of each matching node according to the von Neumann diagram, and each matching node is responsible for the data sending and receiving of the corresponding simulation data management region;

[0030] The client node sends a subscription request for the region of interest data to any matching node, and the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management region overlaps with the region of interest is set as the main matching node of the corresponding client node;

[0031] The client node publishes simulation data to the corresponding main matching node, the main matching node notifies the main matching node of the relevant subscription client node according to the subscription record, and the main matching node sends the simulation data to all relevant subscription client nodes.

[0032] A computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the specific steps of the above-mentioned large-scale joint simulation interest management method based on the Von Neumann diagram when executing the computer program.

[0033] A computer readable storage medium, which stores a computer program, the computer program implements the specific steps of the above-mentioned large-scale joint simulation interest management method based on the Von Neumann diagram when executed by a processor.

[0034] The above-mentioned large-scale joint simulation interest management method and system based on the Von Neumann diagram, by configuring the matching node of the first access system as the network management node, the network management node determines whether the matching node to be accessed can join the system, and generates the Von Neumann diagram based on the distribution information of all matching nodes that have joined the system, and divides the simulation data management area of each matching node according to the Von Neumann diagram, each matching node is responsible for the data sending and receiving of the corresponding simulation data management area, the client node sends a subscription request of the interested area data to any matching node, the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management area overlaps with the interested area is set as the master matching node of the corresponding client node, when the client node publishes the simulation data to the corresponding master matching node, the master matching node notifies the master matching node of the related subscription client node according to the subscription record, and the master matching node sends the simulation data to all related subscription client nodes. By using the method, the redundant simulation data based on network transmission is reduced, and the transmission interaction relationship of the simulation data is optimized, thereby improving the communication efficiency of large-scale distributed simulation and making the operation more smooth. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a flowchart of the large-scale joint simulation interest management method based on the Von Neumann diagram in one embodiment;

[0036] Figure 2 It is a specific step flowchart of the method in one embodiment;

[0037] Figure 3 It is an internal structure diagram of the computer device in one embodiment. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application.

[0039] A large amount of simulation data needs to be transmitted in the process of large-scale distributed simulation system running. The traditional C / S architecture communication mode cannot meet the needs of dynamic expansion of node scale and massive data network transmission. P2P (peer-to-peer) as a decentralized communication mode can obtain better scalability and meet the needs of larger-scale data transmission, so it is more suitable for solving the problems of node expansion and data communication in large-scale distributed simulation.

[0040] Based on the interest management technology to reduce the transmission of redundant data in the network and the operation link, it is very important to improve the overall performance of the distributed simulation system in large-scale application scenarios. It can be used for modeling and simulation of complex systems, and has important application prospects in many civil fields, such as large-scale online games, multi-robot area coverage, urban traffic planning and logistics management.

[0041] The publish / subscribe mode is often used in distributed simulation system to realize the sending and receiving of simulation data. In this interaction mode, first, the user declares the area of interest (AOI) to indicate the information he needs, and then the data sender matches the data with the previous subscription and delivers it to the relevant subscribers.

[0042] The traditional method divides the virtual network space into regular grid cells, and the user receives information by subscribing to the cells intersecting with his area of interest. However, the coverage range of such messages is often larger than the actual needs of the user, resulting in redundant data transmission. If the grid size is reduced to improve the accuracy of subscription, the cost of maintaining the subscription will increase, wasting a lot of CPU processing resources. Moreover, the traditional interest management scheme generally adopts a static partitioning strategy. When the user node position changes, the system has difficulty in updating its topology structure in time, and when a large number of nodes suddenly appear in a local area, overload may even occur. The traditional publish / subscribe mechanism is mostly based on topics. The data sender and the data receiver deliver the sending data to the receiving party through a specific link channel. However, for spatial attribute information, how to divide the space and associate it with the channel is a problem that must be solved.

[0043] The existing interest management method always adopts a static partitioning method. When the simulation node position changes, it cannot be updated in time, and even overload may occur when a large number of nodes suddenly appear in a certain area. The previous publish / subscribe mechanism based on spatial attributes mostly adopts a grid-based spatial partitioning method, which needs to define a spatial grid and cannot be changed during simulation running. When the number of simulation entities is large or the entities move unpredictably, this method becomes very impractical.

[0044] In view of the above problems, such as Figure 1As shown, a large-scale joint simulation interest management method based on von Neumann diagram is provided, which is implemented in a large-scale joint simulation system including matching nodes, network management nodes and client nodes, and specifically includes the following steps:

[0045] In step S100, the matching node of the first access system is configured as a network management node, the network management node determines whether the matching node to be accessed can join the system, generates a von Neumann diagram based on the distribution information of all matching nodes that have joined the system, divides the simulation data management area of each matching node according to the von Neumann diagram, and each matching node is responsible for sending and receiving data in the corresponding simulation data management area.

[0046] In step S110, the client node sends a subscription request for the interested area data to any matching node, and the matching node receiving the subscription request broadcasts the subscription request, and sets the matching node whose simulation data management area overlaps with the interested area as the main matching node of the corresponding client node.

[0047] In step S120, the client node publishes simulation data to the corresponding main matching node, the main matching node notifies the main matching node of the related subscription client node according to the subscription record, and the main matching node sends the simulation data to all related subscription client nodes.

[0048] In this application, by designing a decentralized interest management permission division strategy and a location attribute-based publishing and subscription method, the redundant simulation data based on network transmission is reduced, and the transmission interaction relationship of simulation data is optimized, thereby improving the communication efficiency of large-scale distributed simulation and making it run more smoothly. In this method, the application of von Neumann diagram space segmentation algorithm in the field of distributed simulation interest management is explored, and an interest management method is proposed, which can adapt to the dynamic change of location information of distributed simulation, and can also play the advantages of dynamic expansion and real-time of decentralized network architecture, that is, the interest management method of multiple service centers coordinated in large-scale distributed client nodes.

[0049] In the embodiment, in the large-scale joint simulation system, according to the different task roles of different nodes in the interest management, the nodes are divided into client nodes, matching nodes, nearby matching nodes, alternative matching nodes, main matching nodes, gateway nodes and areas of interest. The client nodes are simulation nodes responsible for publishing or subscribing data. The matching nodes are the management nodes of the Von Neumann area, responsible for recording subscription information and matching the data to be published with potential subscribers. The nearby matching nodes are the matching nodes around the given matching node of the Von Neumann area. The alternative matching nodes are nodes that have the ability and accessibility to be new matching nodes. The main matching nodes are the matching nodes whose area covers the center of the given subscription area. The gateway nodes are the first matching nodes found in the distributed simulation and maintain a record of alternative matching nodes. The area of interest (AOI) is the area corresponding to the data subscribed by the client node.

[0050] In the embodiment, all publishing / subscription requests are sent from data generators or receivers (i.e., publishers and subscribers) to the interest matching nodes, and the main matching nodes save the requests and match the published data with potential subscribers. The main responsibility of the main matching nodes is to record the requests and match the published messages with potential subscribers interested in the messages (i.e., perform interest management between publishers and subscribers).

[0051] In the embodiment, in order to ensure the scalability of large-scale distributed simulation and maintain effective data communication, the method starts from two aspects: one is that each client node (i.e., the node responsible for sending or receiving data) maintains an area of interest (AOI) describing the data visibility of itself, and the other is to limit the data transmission only between the nearby nodes of the AOI to limit the consumption of bandwidth resources.

[0052] In step S100, initialization is first performed, and the first matching node added to the distributed simulation system is the network management node of the entire simulation system. After that, the matching nodes are successively added to the system, and each added matching node sends a join request to the network management, and then the gateway decides whether the matching node to be added can be an alternative matching node, and if so, the information of the matching node is recorded.

[0053] In the embodiment, after all the matching nodes are added to the simulation system, the simulation data management areas of the matching nodes currently added to the system are divided according to the von Neumann diagram by the network management node, including: dividing the virtual space into a plurality of initial simulation data management areas according to the number of all the matching nodes in the system by using the von Neumann diagram, so that each matching node is the nearest node to all the nodes in the region where the matching node is located, and then optimizing the region shape and size of each initial simulation data management area according to the capacity limit of each matching node to obtain an optimized simulation data management area adapted to the node load, constructing a tree structure covering all the neighbors of the region of interest of the matching node based on the optimized simulation data management area and combining the extended tree rule, and finally realizing data transmission according to the optimized simulation data management area and the tree structure.

[0054] In order to perform publish / subscribe more scalable based on multiple interest matching nodes (in a server cluster or P2P overlay), a region division method between nodes needs to be designed. For example, the subscription record must be divided among several interest matching nodes, and the publish request needs to be efficiently routed and distributed to each interest matching node. The division or partition of the spatial domain can be done in many different ways, for example, regular grid, hexagon, bar, triangle, quadtree or von Neumann diagram.

[0055] In the embodiment, the n matching nodes in the space are divided into n regions using the von Neumann diagram, and each node is the nearest node to all the nodes in the region where the node is located. In order to send data within the interest area, a tree structure also needs to be constructed based on the von Neumann diagram.

[0056] Specifically, the simulation data management area is obtained by partitioning based on the von Neumann diagram, and the region shape and size can be continuously adjusted, so that the number of clients in the region follows the capacity of the matching node, which can effectively solve the problems of system scalability and load balancing. The interest management based on the von Neumann diagram divides the virtual space into multiple regions, relies on the existing consistency of the client for management, and fully considers the problems of client heterogeneity and client loss. Using the von Neumann-based interest management allows the client and the server to be integrated in the same framework, realizing seamless and scalable connection between the client and the server.

[0057] Further, in the von Neumann diagram-based interest management, a direct sending scheme is adopted, and the node to be added to the simulation system is taken as a target node, which can establish a connection with a neighbor node to directly communicate, the basic idea of which is to construct a multicast tree covering all the AOI neighbors of the target node, data is transmitted along the branches, and the transmission of redundant data is greatly reduced.

[0058] In the embodiment, the management area is based on the optimized simulation data, and a tree structure covering all the matching nodes of interest is constructed according to the extended tree rule, including: for each matching node, all the neighboring matching nodes of the matching node are excluded from the constructed tree structure, and then the nodes overlapping the region of interest of the target node are screened out, the node with the minimum distance to the target node is calculated from the neighboring matching nodes of the screened nodes, if the node with the minimum distance is the current matching node, the screened node is determined as the child node of the current matching node, and the above process is repeated until all the matching nodes of interest are included in the tree structure, and it is ensured that there is no case that two nodes in the extended tree share the same parent node.

[0059] Specifically, in order to prevent transmission redundancy, two nodes in the extended tree are not allowed to select the same node as a child node. The child node selection process is as follows: x.N: the neighboring node of node x; x.P: the parent node of node x; r.AN: the AOI neighboring node of node r; min_dist(S, r): the minimum value of the Euclidean distance between all nodes in the node set S and node r. Node x selects a child node and transmits the data of node r, that is, the child node y selects node x as its parent node, y.N: the neighboring node of node y. For each y∈((x.N-x.P-r.N)∩r.AN), take z as the node with the minimum distance between y.N and node r, if z is the same node as x, y can be the child node of node x.

[0060] In the embodiment, in the large-scale joint simulation interest management process, if a matching node does not receive any subscription request within a preset time period, the matching node automatically leaves the system, and the network management node adjusts the simulation data management area of the neighboring matching node of the matching node that has left the system.

[0061] Specifically, if there is no subscription information that needs to be managed by a certain matcher within a period of time, the matcher node automatically leaves the distributed simulation system. The neighboring matcher of the matcher node that has left adjusts its Voronoi region. If the matching node that has left repeats the above joining process, it requests to rejoin as a candidate matcher node.

[0062] In the embodiment, each matching node is the unique authoritative node of the simulation data management area where it is located, and each client node needs to register its subscription information with at least one matching node before it can receive data. At the same time, in order to ensure that each subscription request is processed by only one matching node, the matching node covering the center point of the subscription information region can be the main matching node of the subscription information.

[0063] In step S110, the newly joined client node notifies an existing matching node of its subscribed region of interest (ROI). The information of this matching node can be obtained during prior login. If no matching node exists, the node can directly notify the gateway node (when the simulation begins, no matching node has joined the simulation, and no gateway has been registered; the client node joins first and needs to register a matching node, but there is no matching node available, so it registers with the gateway). The matching node receiving this notification broadcasts the subscription information to find the matching node corresponding to the subscribed ROI as the primary matching node. It then notifies the newly joined client node of the relevant information of the primary matching node. Next, the client node requests the subscribed simulation data from its primary matching node.

[0064] In this embodiment, when the matching node that overlaps with the simulation data management area and the area of ​​interest is set as the main matching node of the corresponding client node: the matching node corresponding to at least one simulation data management area covered by the area of ​​interest is taken as the main matching node. When the area of ​​interest covers multiple simulation data management areas, the main matching node pushes the order request to the adjacent matching node to back up the order information.

[0065] In step S120, when a client node publishes simulation data, the main matcher node receives the data and checks whether the published data matches its corresponding subscription information record. If a match is found, the main matcher node that subscribed to the region of interest associated with the published simulation data is notified, and the main matcher node then sends the data to the corresponding client node.

[0066] In this embodiment, after a client node publishes simulation data to the corresponding main matching node, if the region of interest associated with the published simulation data overlaps with the simulation data management region managed by a neighboring matching node of the main matching node of the relevant ordering client node, the simulation data is pushed sequentially to all neighboring matching nodes whose simulation data management regions overlap with the region of interest associated with the published data. The neighboring matching nodes then publish the simulation data to the corresponding client node, ensuring that the simulation data is sent only once to a particular client node.

[0067] Specifically, each matching node and its neighboring matching nodes, when sending data, the data is first sent to the neighboring matching nodes, and then pushed forward to the neighboring matching nodes of that neighboring matching node, ensuring that all neighboring matching nodes of the region of interest receive the data.

[0068] like Figure 2 The diagram shown illustrates the complete steps of this method.

[0069] In this embodiment, a general space publish / subscribe mechanism is proposed by combining space proactive query and space broadcast. In the space publish / subscribe (SPS) mechanism, each client node in the distributed simulation system can specify a publish space and a subscribe space (i.e. area of interest). The client node can send data to its specified publish space, and then the data will be delivered to all nodes in the subscribe space that overlaps with the publish space. The publish / subscribe spaces are also allowed to update as the nodes move (but it is noted that the new space should overlap with the previous space mostly).

[0070] Further, once the subscription relationship is declared, the data receiver in the SPS does not need to continuously query at different intervals to receive data updates. The data receiver does not need to resend any request unless the subscription area has changed. Therefore, the delay between the occurrence of the update and the discovery of the update is reduced. Take the user location as an example. In the space query method, the mobile user node needs to register its new location on the logger every certain period of time, and other users must continuously query the updated location within the AOI neighborhood.

[0071] In this embodiment, each client node only needs to subscribe to the data within its AOI, and broadcasts its location to all other nodes whenever its location changes. It is necessary to ensure that the subscribed area also changes as the node corresponding entity moves, and each node can send according to the updated subscription area when the location update occurs.

[0072] In this embodiment, when the center point of the subscribed area of interest moves from the boundary of one matcher node into the area of another matcher node, an explicit ownership transfer is required. The new primary matcher node needs to exchange data with the old matcher node to ensure that the change of the matcher node has no effect on the client node.

[0073] In this embodiment, in the process of data sending and receiving, when a data publication occurs, the data is first sent from the client node to its home matcher node, and then it sends the data after checking the list of subscribers. When the sending data region (i.e. the region of interest) is outside the home matcher node region, the publication information is directly given to the adjacent matcher node until all affected matchers are notified. This ensures that when the subscription information is correctly maintained by each matcher node, each publication information is delivered to all possible subscribers. Although a subscription region can cover multiple von Neumann regions, each subscription information has only one home matcher node, which is responsible for delivering all related publication information to the subscription node. In order to ensure that the subscription node can still receive updated data from other area matcher nodes, the subscription information needs to be replicated in all other non-home matcher nodes with overlapping regions. In this case, the received data will have temporary inconsistency, but will eventually converge. The matcher node with replicated subscription information checks whether there is publication data in the area it is responsible for. These publication data will be pushed to the subscription node, i.e. the home matcher node of a client node, which will notify the subscription node of the publication data information.

[0074] Specifically, in this method, data is sent to all adjacent matcher nodes, and the adjacent matcher nodes at the edge are responsible for checking new adjacent matcher nodes. At the same time, the connection to the new matcher node is in the form of a notification, and the node without coverage of AOI is disconnected.

[0075] In the above large-scale joint simulation interest management method based on von Neumann diagram, the distributed simulation transmission architecture based on von Neumann diagram has better distribution characteristics, low latency, high loan utilization, etc. This method can realize spatial dynamic segmentation during simulation running, and dynamically adjust the spatial segmentation according to the density or distribution change of the entity, which makes the publication and subscription based on spatial information more adaptive and flexible. In this method, the following characteristics are realized based on the point-to-point (P2P) architecture: the client node can be added or removed at any time, and dynamic load balancing is supported. Relying on the von Neumann diagram for region division, neighbor discovery and spatial query can be completed in a limited time, that is, the query time will not be prolonged with the increase of the number of nodes.

[0076] It should be understood that, although Figure 1 The steps in the flowchart of the method shown in FIG. 8 are shown in sequence according to the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, Figure 1At least one of the steps in the method can comprise a plurality of sub-steps or a plurality of stages, which sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the order of the sub-steps or stages is not necessarily sequential, but can be performed alternately or in rotation with other steps or sub-steps or stages of other steps.

[0077] In one embodiment, a large-scale joint simulation interest management system based on von Neumann diagram is provided, which comprises a matching node, a network management node and a client node, and in the system:

[0078] The matching node accessing the system for the first time is configured as the network management node, the network management node determines whether the matching node to be accessed can join the system, generates a von Neumann diagram based on the distribution information of all the matching nodes that have joined the system, and divides the simulation data management area of each matching node according to the von Neumann diagram, and each matching node is responsible for the sending and receiving of data in the corresponding simulation data management area.

[0079] The client node sends a subscription request for the area of interest data to any matching node, the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management area overlaps with the area of interest is set as the primary matching node of the corresponding client node.

[0080] The client node publishes simulation data to the corresponding primary matching node, the primary matching node notifies the primary matching nodes of the related subscription client nodes according to the subscription record, and the primary matching node sends the simulation data to all the related subscription client nodes.

[0081] Specifically, the network management node is assumed by the matching node accessing the system for the first time, and is used to determine whether the matching node to be accessed can join the system, generate a von Neumann diagram based on the distribution information of the accessed matching nodes, and divide the simulation data management area of each matching node according to the von Neumann diagram.

[0082] Specifically, the matching node is responsible for the sending and receiving of data in the corresponding simulation data management area, broadcasts the AOI subscription request of the client node, sets itself as the primary matching node of the client node when the simulation data management area overlaps with the AOI, and notifies the related primary matching nodes according to the subscription record and forwards the simulation data.

[0083] Specifically, it is used to send a subscription request for the AOI data to any matching node, publish simulation data to the corresponding primary matching node, and receive the simulation data forwarded by the primary matching node.

[0084] The specific definition of the large-scale joint simulation interest management system based on the Von Neumann diagram can refer to the definition of the large-scale joint simulation interest management method based on the Von Neumann diagram in the above, which will not be repeated here. Each module in the large-scale joint simulation interest management system based on the Von Neumann diagram can be realized by software, hardware, and a combination thereof, in whole or in part. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so that the processor calls and executes the operations corresponding to each of the above modules.

[0085] In one embodiment, a computer device, which can be a terminal, has an internal structure diagram as shown in Figure 3 The computer device includes a processor, a memory, a network interface, a display screen, and an input device connected by a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement a large-scale joint simulation interest management method based on the Von Neumann diagram. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball, or touchpad provided on the shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0086] Those skilled in the art can understand that Figure 3 The structure shown in the above is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0087] In one embodiment, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following steps:

[0088] In step S100, the matching node of the first access system is configured as a network management node, which decides whether the matching node to be accessed can join the system, and generates a Von Neumann diagram based on the distribution information of all matching nodes that have joined the system, divides the simulation data management area of each matching node according to the Von Neumann diagram, and is responsible for data transmission and reception in the corresponding simulation data management area.

[0089] Step S110, the client node sends a subscription request of the interested area data to any matching node, the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management area overlaps with the interested area is set as the main matching node of the corresponding client node.

[0090] Step S120, the client node publishes simulation data to the corresponding main matching node, the main matching node informs the main matching nodes of the relevant subscription client nodes according to the subscription record, and the main matching node sends the simulation data to all the relevant subscription client nodes.

[0091] In one embodiment, a computer readable storage medium is provided, and the computer program is stored on the computer readable storage medium. The computer program is executed by a processor to implement the following steps:

[0092] Step S100, the matching node accessing the first system is configured as a network management node, the network management node determines whether the matching node to be accessed can join the system, generates a von Neumann diagram based on the distribution information of all the matching nodes having joined the system, and divides the simulation data management areas of the matching nodes according to the von Neumann diagram, so that each matching node is responsible for the sending and receiving of data in the corresponding simulation data management area.

[0093] Step S110, the client node sends a subscription request of the interested area data to any matching node, the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management area overlaps with the interested area is set as the main matching node of the corresponding client node.

[0094] Step S120, the client node publishes simulation data to the corresponding main matching node, the main matching node informs the main matching nodes of the relevant subscription client nodes according to the subscription record, and the main matching node sends the simulation data to all the relevant subscription client nodes.

[0095] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, databases, or other media in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0096] The technical features of the above embodiments can be combined in any way. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0097] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for managing interests in large-scale co-simulation based on a von Neumann diagram, characterized in that, The method is implemented in a large-scale joint simulation system comprising matching nodes, network management nodes and client nodes, and the method comprises: configuring a matching node of a first access system as a network management node, determining whether a matching node to be accessed can join the system by the network management node, generating a von Neumann diagram based on distribution information of all matching nodes that have currently joined the system, and dividing simulation data management areas of the matching nodes according to the von Neumann diagram, wherein the simulation data management areas of the matching nodes are responsible for data transmission and reception of the corresponding simulation data management areas, and wherein the dividing of the simulation data management areas of the matching nodes according to the von Neumann diagram comprises: dividing a virtual space into a plurality of initial simulation data management areas corresponding to the number of all matching nodes in the system by using the von Neumann diagram, so that each matching node is the nearest node to all nodes in the region where the matching node is located, optimizing the region shape and size of each initial simulation data management area according to the capacity limit of each matching node to obtain an optimized simulation data management area that adapts to the load of the node, constructing a tree structure covering neighbors of the region of interest of all matching nodes based on the optimized simulation data management area and in combination with an extended tree rule, and realizing data transmission according to the optimized simulation data management area and the tree structure; the client node sends a subscription request for the region of interest data to any matching node, and the matching node receiving the subscription request broadcasts the subscription request, and the matching node whose simulation data management area overlaps with the region of interest is set as the primary matching node of the corresponding client node; the client node publishes simulation data to the corresponding primary matching node, the primary matching node notifies the primary matching nodes of the relevant subscription client nodes according to the subscription record, and the primary matching node sends the simulation data to all relevant subscription client nodes.

2. The von Neumann diagram based large-scale co-simulation interest management method according to claim 1, characterized in that, constructing a tree structure covering neighbors of the region of interest of all matching nodes based on the optimized simulation data management area and in combination with an extended tree rule comprises: for each matching node, excluding the parent node and the target node from all neighboring matching nodes of the matching node, and then screening out nodes overlapping with the region of interest of the target node; calculating the node with the smallest distance to the target node among the neighboring matching nodes of the screened nodes, and if the node with the smallest distance is the current matching node, the screened node is determined as the child node of the current matching node; repeating the above process until all neighbors of the region of interest of all matching nodes are included in the tree structure, and ensuring that there is no case where two nodes in the extended tree share the same parent node.

3. The von Neumann diagram-based management method of interest in large-scale joint simulation according to claim 1 or 2, characterized in that, when setting the matching node whose simulation data management area overlaps with the region of interest as the primary matching node of the corresponding client node: the matching node corresponding to at least one simulation data management area covered by the region of interest is taken as the primary matching node; when the region of interest covers multiple simulation data management areas, the primary matching node pushes the subscription request to the neighboring matching node to backup subscription information.

4. The von Neumann diagram based large-scale joint simulation interest management method according to claim 3, characterized in that, after the client node publishes simulation data to the corresponding primary matching node; If the region of interest associated with the published simulation data overlaps with the simulation data management area of the neighboring matcher node of the master matcher node of the relevant subscription client node, the simulation data is pushed to all the neighboring matcher nodes whose simulation data management areas overlap with the region of interest associated with the published simulation data, and the simulation data is published to the corresponding client node by the neighboring matcher node.

5. The von Neumann diagram based large-scale joint simulation interest management method according to claim 4, characterized in that, In the large-scale joint simulation interest management process, if a matcher node does not receive any subscription request within a preset time period, the matcher node automatically leaves the system. Meanwhile, the network management node adjusts the simulation data management area of the neighboring matcher node responsible for the matcher node that leaves the system.

6. The von Neumann diagram based large-scale joint simulation interest management method according to claim 5, characterized in that, When the client node sends a subscription request for the region of interest data to any matcher node, if there is no matcher node in the system, the client node directly sends a subscription request to the gateway node.

7. A large-scale federated simulation interest management system based on a von Neumann diagram, characterized in that, The system includes matcher nodes, network management nodes, and client nodes, and in the system: The first matcher node that accesses the system is configured as a network management node, which determines whether the matcher node to be accessed can join the system, generates a von Neumann diagram based on the distribution information of all the matcher nodes that have joined the system, and divides the simulation data management areas of the matcher nodes according to the von Neumann diagram, and each matcher node is responsible for the data transmission and reception of the corresponding simulation data management area. According to the von Neumann diagram, the simulation data management areas of the matcher nodes are divided, which includes: using the von Neumann diagram, dividing the virtual space into a corresponding number of initial simulation data management areas according to the number of all matcher nodes in the system, so that each matcher node is the nearest node to all nodes in its region, optimizing the region shape and size of each initial simulation data management area according to the capacity limit of each matcher node, to obtain an optimized simulation data management area that adapts to the node load, and based on the optimized simulation data management area, combined with the expansion tree rule, a tree structure covering the neighbors of the region of interest of all matcher nodes is constructed, and the data transmission is realized according to the optimized simulation data management area and the tree structure. The client node sends a subscription request for the region of interest data to any matcher node, and the matcher node that receives the subscription request broadcasts the subscription request, and the matcher node whose simulation data management area overlaps with the region of interest is set as the master matcher node of the corresponding client node. The client node publishes simulation data to the corresponding master matcher node, and the master matcher node notifies the master matcher node of the relevant subscription client node according to the subscription record, and the master matcher node sends the simulation data to all relevant subscription client nodes.

Images