18 results about "Load balanced routing" patented technology

The invention relates to the technical field of unmanned aerial vehicle group networks, in particular to a dynamic clustering routing method for a high-dynamic unmanned aerial vehicle group network, which is characterized by comprising the following steps of: dividing the whole unmanned aerial vehicle group network into a centralized-distributed hierarchical domain architecture, dividing the network into a plurality of clusters, and enabling each cluster to consist of a cluster head and a plurality of cluster members, the cluster head can be recombined into a higher-level cluster; according to the method, a layered and domain-divided network architecture is provided, cluster heads are dynamically elected in combination with a grey wolf optimization algorithm, real-time self-adaptive clustering and load balancing routing of the unmanned aerial vehicle group in a high-speed mobile environment are realized through a double-cluster-head switching and reselection mechanism, the cluster heads and paths are ensured to be optimal, routing is optimized through a genetic algorithm, relay forwarding is reduced, and the routing efficiency is improved. The communication overhead is reduced, the service life of the network is prolonged, the robustness of the system is enhanced by the alpha-beta-delta leader model and the iterative surrounding strategy, the time delay is reduced and the transmission success rate is improved by the fuzzy clustering, and the real-time reliable communication requirement in a high dynamic scene is met.

The invention provides a neuromorphic computing system and a neuromorphic computing power extension routing method, the computing system comprises a plurality of neuromorphic computing chips, and each computing chip is integrated with a plurality of computing cores; the hierarchical distributed routing network is used for connecting a plurality of computing chips, and the routing network comprises an on-chip routing layer which is arranged in a single computing chip and is used for realizing local interconnection and data exchange between computing cores; the inter-chip routing layer is bridged among the plurality of computing chips and is used for realizing global interconnection and computing power expansion among the computing chips; the routing control module is coupled with a routing network and is used for dynamically selecting and executing a shortest path routing strategy, a load balancing routing strategy or an event-driven routing strategy according to network state information acquired in real time; and the system resource management module is used for performing pooling management and dynamic scheduling on the computing resources of the computing core and the communication resources of the routing network.

The invention relates to the field of MCP protocol integrated management, in particular to an MCP integrated management system and method supporting multi-task collaboration, and the method comprises the steps: completing the deployment of one or more service instances according to the communication mode of an MCP Server through employing a corresponding deployment strategy, and registering the metadata of the MCP Server and the information of the service instances into a resource library; creating a personal cooperation space and a team cooperation space, and configuring role permissions for members of the team cooperation space; generating a workflow definition containing task execution logic; according to the workflow definition, creating and executing a corresponding workflow instance, and feeding back an execution state in real time; receiving a calling request for the managed MCP Server, and carrying out load balancing routing and protocol conversion to realize seamless calling of the third-party MCP service; the requirements in different scenes are met, and the flexibility and efficiency of cooperation are improved.

The invention discloses a time sequence data partition storage method for a high-frequency acquisition system, relates to the technical field of data storage management, and is used for solving the problem of dynamic unbalance of a high-frequency write-in structure. According to the method, a time sequence data dynamic storage mechanism based on write-in pressure sensing and partition structure linkage is constructed, rapid response and structure self-adaptive adjustment of burst write-in loads in a high-frequency acquisition system are achieved, the write-in overload state is recognized by constructing a sliding write-in rate sequence and a write-in pressure factor, and therefore the write-in pressure sensing and partition structure self-adaptive adjustment of the high-frequency acquisition system is achieved. The method comprises the following steps: dynamically triggering a structure splitting operation of a main partition, generating a sub-partition with a time boundary and an independent buffer path, judging and realizing a load balancing route of a write-in request by adopting timestamp mapping and a buffer area utilization rate, generating a structure merging evaluation factor in combination with write-in activeness and delay statistics, and dynamically adjusting a partition structure according to trend change, and the write-in structure and the system load are cooperatively matched, so that the structural elasticity and the write-in efficiency of the system in a high-concurrency scene are improved.

The invention discloses a link congestion control load balancing routing method based on routing domain division. The method belongs to the technical field of low-orbit satellite routing switching application. The problem that a traditional routing algorithm is high in complexity and slow in response time under the condition of a large network scale is solved. A giant constellation is divided into non-overlapping routing domains, global routing is divided into intra-domain local routing calculation and cross-domain routing interaction, on-satellite computing power consumption is reduced, and the method is adaptive to a large-scale satellite network. Meanwhile, a traditional distributed routing algorithm relates to signaling flooding when the state of a satellite link is changed, the signaling overhead is huge in a large-scale satellite network, local response is carried out in the congestion process by means of a sub-domain architecture and inter-satellite link availability prediction, and the signaling overhead of the whole network is reduced, so that the availability of the routing algorithm is improved.

This invention provides a method and system for vehicular data transmission based on spatiotemporal autonomous load balancing routing, belonging to the field of intelligent network routing technology. The method includes: sending vehicle driving data to a non-congested network relay device; filtering the vehicle driving data by the non-congested network relay device to obtain a reference trajectory path; sensing the location and storage information of neighboring network relay devices and outputting a set of candidate relay devices; calculating the spatiotemporal parameter set of the candidate relay device set, and obtaining the comprehensive data transmission capability of the candidate relay device set based on weights and the spatiotemporal parameter set; updating network status data and the congestion status of network relay devices; and arranging the data transmission order in ascending order according to the remaining lifetime, transmitting the vehicle driving data to the data transmission destination. This invention employs a weighted and spatiotemporal autonomous load balancing routing method to comprehensively evaluate the data transmission capability of network relay devices and select the optimal next-hop network relay device for the data.

A communication method includes: establishing, in response to an SCTP connection request transmitted by a base station device, an SCTP connection with the base station device; receiving a service setup request transmitted by the base station device based on the SCTP connection after the SCTP connection is established; encapsulating request data included in the service setup request as a request message; querying, according to pre-configured AMF domain name information, for an AMF network address corresponding to the AMF domain name information; transmitting the request message to a target AMF corresponding to the AMF network address; encapsulating, based on an SCTP protocol, a response message returned by the target AMF for the request message to obtain a response result; and transmitting the response result to the base station device.

The present invention discloses a heuristic traffic scheduling method based on delay tolerance, which relates to the field of time-sensitive network technology and includes the following steps: S1, calculating load balancing routes, and distributing traffic relatively evenly to all available links through load balancing strategies to avoid network congestion and improve overall bandwidth utilization; S2, calculating delay-tolerant scheduling, setting scheduling strategies, and tolerating delays in intermediate nodes so that larger-scale traffic can be scheduled under the same conditions; S3, calculating compression, and optimizing the use of network bandwidth by reducing the number of guard bands, thereby improving the utilization efficiency of network resources. The present invention improves traffic schedulability. Under the same scheduling conditions, a larger number of time-sensitive flows can be scheduled through a delay-tolerant mechanism; at the same time, bandwidth utilization is also improved. Through load balancing routes and compression algorithms, the system can more reasonably allocate network bandwidth, avoiding bandwidth waste.

The invention provides a multi-physical WAN port intelligent load balancing routing method and device, and the method comprises the steps: achieving the division of a load balancing domain through the creation and configuration of WAN connection; the load balancing strategy of the WAN connection is configured and takes effect to a Linux kernel protocol; and after the LAN port receives the data packet, a certain physical WAN port is selected as a data packet forwarding outlet for forwarding according to the load balancing domain and the load balancing strategy. According to the invention, for the gateway equipment with multiple physical WAN ports, the data traffic on the LAN side can be intelligently distributed and scheduled to different physical WAN ports according to the specified load balancing rule.

A method, system, device and medium for a trusted load balancing routing in a low-earth orbit satellite network. The method includes: constructing a communication scenario for the low-earth orbit satellite network; modeling the quality of service constraints for the trusted load balancing routing in the low-earth orbit satellite; modeling the objective function for the trusted load balancing routing in the low-earth orbit satellite; establishing a fully distributed multi-agent architecture; establishing a D3QN deep reinforcement learning training model in the fully distributed multi-agent architecture, setting the state, action and reward of the training model, and conducting training for the routing decision model; designing a trusted load balancing routing strategy for the low-earth orbit satellite network; the system, device and medium are used to implement a method for a trusted load balancing routing in a low-earth orbit satellite network; the present invention solves the problems of trusted routing and load balancing routing in the low-earth orbit satellite scenario, and can reduce the maximum queue utilization rate of the network while meeting the trust requirements and latency-sensitive requirements of the services, and balance the network load.

The present disclosure relates to an inference system and method for large models. The system comprises a routing component, a pre-filling component, a decoding service component deployed independently of the pre-filling component, and a communication component between the pre-filling component and the decoding service component. The routing component routes long data processing requests to the pre-filling component for pre-filling processing based on token information of the current data processing request, routes short data processing requests to the decoding service component for decoding processing, and performs load balancing routing based on load states of multiple inference capabilities in the pre-filling component or the decoding service component.

This invention discloses a time-series data partitioning and storage method for high-frequency acquisition systems, relating to the field of data storage management technology, and is used to solve the problem of dynamic imbalance in high-frequency write structures. This invention achieves rapid response and adaptive structural adjustment to sudden write loads in high-frequency acquisition systems by constructing a dynamic time-series data storage mechanism based on write pressure sensing and partition structure linkage. By constructing a sliding write rate sequence and a write pressure factor, it identifies write overload states, dynamically triggers the structural splitting operation of the main partition, generates sub-partitions with time boundaries and independent buffer paths, uses timestamp mapping and buffer utilization judgment to achieve load-balanced routing of write requests, and combines write activity and latency statistics to generate a structure merging evaluation factor. The partition structure is dynamically adjusted according to trend changes, enabling the write structure to match the system load, thereby improving the system's structural resilience and write efficiency in high-concurrency scenarios.

The application discloses a routing method, device and equipment of a time-sensitive network (TSN), a medium and a product, and the method comprises the following steps: acquiring a network topology structure of a TSN network, determining time-triggered (TT) flows in the network topology structure and basic transmission characteristic parameters of the TT flows, and the network topology structure comprises switches, sending terminals and receiving terminals; determining a link load rate of data links contained in the TSN network according to the basic transmission characteristic parameters; performing TSN load balancing routing on the TT flows based on the link load rate to obtain an optimal path of the TSN network. The data link load rate is determined through the basic transmission characteristic parameters of the TT flows, and then routing planning is performed based on the load rate, so that the TT flows are reasonably distributed in multiple shortest paths, TSN network link load balancing is realized, the situation that multiple TT flows concentrate on occupying the same data link is avoided, and transmission conflicts are reduced through balanced link load.

The invention relates to the technical field of network communication, and discloses a load balancing routing method, device and equipment and a storage medium, and the method comprises the steps: obtaining target network situation information corresponding to other nodes in a mobile ad hoc network according to a preset sensing period; determining a comprehensive load index based on the target network situation information; determining a target routing path through the target network situation information and the comprehensive load index; and sending the service data to the destination node according to the target routing path. In the mobile ad hoc network, each node senses the target network situation information of the surrounding nodes in a distributed manner, the target routing path is determined according to the target network situation information and the comprehensive load index, and the data packet is sent to the target node according to the target routing path, so that load balancing of each node on the target routing path is realized; unbalanced service distribution is avoided, and the system packet loss rate is reduced, so that the end-to-end time delay of the system is reduced; and through effective balancing of each node, the throughput of the whole system is improved.

The invention discloses a satellite optical network load balancing routing and wavelength allocation method, and relates to the field of satellite optical communication. The method comprises the following steps: establishing a dynamic simulation environment of a satellite optical network; defining a node as an agent, for an arriving service request, starting from a source node agent, taking a current node on a path as an active agent, and selecting a next hop node through deep reinforcement learning until a destination node is reached, thereby forming a candidate path and a to-be-allocated wavelength set; and calculating an allocation metric value for each wavelength to be allocated, and selecting the wavelength with the maximum allocation metric value as the optimal wavelength to be allocated to the current service. According to the method, the use condition of the wavelength in the satellite optical network is sensed through the intelligent agent, the path bypassing the blocking area is selected to avoid cascade blocking, the problem of local congestion on the satellite caused by unbalanced distribution of ground services in geography and time can be effectively solved, the service blocking rate is remarkably reduced, and the network load balancing degree is achieved.

The invention relates to a load balancing routing method and system based on dynamic region segmentation and a medium, the method is operated on double time scales and comprises a dynamic region segmentation stage of a long time slot scale and a hybrid routing stage of a short time slot scale, and the method comprises the following steps: in the dynamic region segmentation stage, based on a dynamic region segmentation algorithm, carrying out hybrid routing on the dynamic region based on the hybrid routing stage; dynamically adjusting cluster attribution of the satellite nodes according to real-time loads and historical flow information of the satellite nodes; in the hybrid routing stage, a head node of a current cluster selects a next hop cluster for a data stream in a distributed mode according to the real-time geographic position and the load state of a neighbor cluster; and the head node of the current cluster constructs a low-load routing path in the cluster for the data flow in a centralized mode based on a residual topology method so as to maximize the throughput of the system. According to the invention, the load balancing routing of the data transmission stream of the large-scale satellite network is realized, and the overall load balancing effect of the network and the reliability of the transmission stream are improved.

The invention discloses a routing method, device and equipment of a time sensitive network, a medium and a product, and the method comprises the steps: obtaining a network topology structure of a time sensitive network TSN, and determining a TT flow triggered with time and a basic transmission characteristic parameter of the TT flow in the network topology structure, the network topology structure comprising a switch, a sending terminal and a receiving terminal; determining a link load rate of a data link included in the TSN network according to the basic transmission characteristic parameter; and performing TSN load balancing routing on the TT flow based on the link load rate to obtain an optimal path of the TSN network. According to the invention, the load rate of the data link is determined through the basic transmission characteristic parameter of the TT flow, and then the route planning is carried out based on the load rate, so that the TT flow is reasonably distributed in a plurality of shortest paths, the load balancing of the TSN network link is realized, the situation that the same data link is occupied by a plurality of TT flows in a centralized manner is avoided, and the transmission conflict is reduced through balancing the link load.

The present disclosure relates to a load balancing routing system and method for large models. The system comprises: a request parsing component that receives a data processing request of a user to obtain token information of the data processing request; a subset segmentation component that divides a token subset from a head of the token information of the received data processing request each time based on a predetermined token number, and generates a hash value of each divided token subset to form one or more hash value sets; a load information collection component that collects, for each destination processing instance, the hash value set and a load amount of data processing requests being processed; a scoring component that scores each destination processing instance based on a matching degree between the hash value set of the current data processing request and each destination processing instance and a load amount of each destination processing instance; and a forwarding component that routes and forwards the current data processing request to a destination processing instance with the highest score or one of destination processing instances with the highest scores.