32 results about "Routing architecture" patented technology

The invention relates to the related technical field of resource scheduling optimization, in particular to a resource scheduling optimization method and system combined with cloud-side collaboration, and the method comprises the steps: an edge computing node carries out the interactive analysis of a computing load, bandwidth fluctuation and a cloud energy consumption center, configures a resource demand and a flow pressure, predicts a bandwidth trend, and sets a flow peak valley; and a lightweight routing architecture is deployed to initialize scheduling, bidirectional optimization is carried out by taking a node as a center, self-adaptive load balancing is carried out, and a cloud edge collaborative resource scheduling strategy is obtained. The technical problem that transmission delay and bandwidth waste are aggravated due to the fact that a resource scheduling strategy lags behind an actual network state due to the fact that network bandwidth change trend prediction accuracy is insufficient and dynamic characteristics of traffic peak values and traffic valley values cannot be accurately captured is solved, a dynamic prediction model is established based on node resource requirements and traffic pressure, and the network bandwidth change trend prediction efficiency is improved. The technical effects of accurately dividing the peak value and valley value intervals of the network traffic, carrying out bidirectional resource allocation by taking the edge node as the center and combining the traffic peak and valley characteristics, and improving the bandwidth utilization rate are achieved.

The invention relates to a deadlock-free core grain routing architecture construction method and device and a processor, and the method comprises the following steps: constructing a multi-data flow network structure in a core grain: constructing an internal subnet which is used for transmitting data in the core grain, does not interact with the outside of the core grain, and comprises at least one routing node and a function module capable of receiving and transmitting data; an outbound subnet is constructed, is used for data transmission in the direction away from the core particles, and comprises at least one routing node, a data transceiving function module and a boundary router; constructing an inbound subnet which is used for data transmission in a core particle direction and comprises at least one routing node, a data transceiving function module and a boundary router; a transmission subnet is constructed, is used for data transmission of which the starting point and the terminal point are other core particles, and comprises at least one routing node, a transmission module and a boundary router; and preventing data transmission from deadlock in the core particles and / or among the core particles according to the multi-data flow network structure through a core particle interaction protocol.

The invention discloses a wafer-level integrated system-oriented network-on-chip routing architecture and a working method thereof, the network-on-chip routing architecture forms a two-dimensional mesh topology structure by a plurality of router nodes, and routers are connected with one another through crisscrossed links in four directions; the basic unit of the network-on-chip field area routing architecture is a router; according to the system, flexible configuration of routing tables and functional modules of all routers is realized through a configuration packet in a specific format. The system adopts a detection method based on a heartbeat packet, a router actively sends a heartbeat request to a neighbor and waits for a response, and meanwhile, a layered health state monitoring mechanism is realized at an on-chip network field level. According to the mechanism, a local monitor is responsible for summarizing router and link states in a region, and a global monitor integrates information of each region to generate a field-level health report. The configuration and fault detection method provided by the invention has the advantages of flexibility, high efficiency, high expandability, high detection precision and the like, and is suitable for the network-on-chip in a large-scale wafer-level integrated system.

The invention discloses a lightweight routing design method and system suitable for a chain broadband wireless ad hoc network. The system comprises a configuration network node, a definition node routing architecture, a data packet classifier module configured in the node, a delay module, a general module, a neighbor list module, a neighbor information manager module, a delay estimator module, a data packet queue module, a queue processing module, an MAC receiver module and an MAC transmitter module. The method comprises the following steps: calculating queues of delay data packets and general data packets according to data frame structure classification; regularly updating a neighbor list according to the data packets of the adjacency relationship; estimating the time required by local queue transmission; and queuing the delay data and the general data according to the priority, and establishing a forwarding link. According to the invention, the nodes automatically complete path planning and queue ordering in real time according to a routing protocol architecture, unified and centralized scheduling by an upper-layer route is not needed, and network resources are not occupied. According to the method, node resources are utilized to the greatest extent on the basis of ensuring communication delay.

The application provides a roaming communication method and network element under a home routing architecture, a storage medium and a computer program product, and comprises the following steps: a first network element receives a domain name resolution request initiated by a terminal; the domain name resolution request carries first domain name information and first DNS address information; in the case that the first domain name information is domain name information of a specific service, the user is a specific user and / or the first DNS address information is specific DNS address information, the first DNS address information is modified into second DNS address information; the DNS server corresponding to the second DNS address information is located in a visited network with the first network element; the domain name resolution request is forwarded to the DNS server, so that the DNS server resolves the first domain name information, obtains first service server address information of the visited network, and returns the first service server address information to the terminal. The service path can be shortened, and the service time delay can be reduced.

Various implementations described herein refer to a device having a multi-layered logic structure with multiple layers including a first layer and a second layer arranged vertically in a stacked configuration. The device may have a first network that links nodes together in the first layer. The device may have a second network that links the nodes in the first layer together by way of the second layer so as to reduce latency related to data transfer between the nodes.

A multi-mode routing optimization method based on multi-agent deep reinforcement learning belongs to the field of edge computing and network routing, and comprises the following steps: constructing a multi-mode routing architecture, and designing a multi-mode routing node interaction mechanism; constructing a state space, an action space, a space state transition probability and a reward function of the multi-agent; designing a double-layer semantic self-attention mechanism; constructing and training an enhanced multi-agent deep reinforcement learning model; designing a trigger condition of dynamic switching of the intelligent agent operation protocol; and designing a model performance evaluation function and a hyper-parameter tuning strategy. According to the invention, the routing nodes with switchable working modes are designed, so that the routing protocol can be dynamically selected according to the network state; a centralized training and distributed execution framework is utilized, and local and global value networks are combined to realize collaborative optimization of multiple agents; and a double-layer semantic self-attention mechanism is introduced to improve the analysis capability of the heterogeneous network state, so that the routing reliability is ensured, the network throughput is improved, and the transmission delay is reduced.

The invention discloses a satellite-ground integrated routing method suitable for a low-orbit constellation, and belongs to the technical field of satellite communication networks. According to the method, a'satellite-to-ground segmented bearing 'routing architecture is adopted, and inter-satellite routing and satellite-to-ground routing are designed in a decoupling manner and are independent from each other. The inter-satellite routing process adopts a mode of combining'top injection topology and autonomous routing ', initial topology is preset on the ground, satellites autonomously monitor link states and change information is synchronized in the whole network, and fault autonomous recovery and global routing convergence are realized; in the satellite-ground routing process, a mode of'injection on the ground and diffusion in the whole network 'is adopted, and the mapping relation between the gateway station and the landing satellite is quickly synchronized. In the forwarding process, according to whether the data fall to the ground or not, the satellite-ground routing tables are inquired to obtain destination satellite addresses, and then the inter-satellite routing tables are inquired to carry out hop-by-hop forwarding. According to the method, the problem of frequent convergence of satellite-ground routing caused by rapid movement of the satellite is effectively solved, the routing dynamics is simplified, and the method has the advantages of being small in protocol overhead, rapid in fault recovery and high in engineering realizability.

The invention discloses an on-chip network hybrid routing implementation method and architecture based on real-time congestion perception, and relates to the technical field of communication of an integrated circuit on-chip system. Comprising the following steps: step 1, creating a hybrid routing architecture; step 2, judging a routing direction by a routing direction calculation module in east, south, west and north input ports by using an NSWE hybrid routing algorithm aiming at a path node; step 3, a routing direction calculation module in a local port judges a routing direction by using a Local hybrid routing algorithm aiming at the source node; according to the invention, the transmission delay of the NoC router is reduced, the influence of congestion on the transmission efficiency is reduced, and the overall performance is improved.

The invention relates to a running method of an embedded code message routing architecture and an architecture system. The running method of the embedded code message routing architecture comprises the following steps: setting at least one routing management module and at least one queue module; the routing management module comprises a routing memory application function and a routing memory sending function; the routing management module applies for a memory from a memory management module through a routing memory application function so as to obtain a memory space for routing message data; and the routing management module also sends the message data needing to be sent to each queue module through a routing memory sending function.

The invention discloses a method and a system for generating a complex and variable-oriented adaptive network. The method comprises the following steps of: constructing a cross-layer routing architecture consisting of a routing layer and an MAC (Media Access Control) layer; after the MAC layer obtains a first-hop neighbor and a second-hop neighbor, respective backbone nodes are selected in a distributed mode, flooding is conducted to the whole network, and declarations of the flooding backbone nodes are grouped and loaded into beacons; the routing layer starts to run on the backbone nodes, performs secondary relay selection on the backbone nodes, notifies the positions of the relay nodes of the whole virtual backbone network, and establishes a local topology table; meanwhile, a zero-interval triggering type cross-layer cooperation flooding algorithm and a constraint flooding algorithm based on fuzzy vision are provided; and finally, selecting an optimal routing path by taking the media occupancy time as a scale for measuring the advantages and disadvantages of the path. According to the invention, the network topology structure can be adaptively, rapidly and automatically adjusted according to the task and node state, the network delay and overhead are reduced, and rapid routing selection and maintenance can be realized on a large scale.

This invention relates to the field of network technology, specifically to a routing architecture and method for an open interconnected network for computing power services. The architecture includes: a network access device; a network output device; a network control center; and a network management device, used to generate a first orchestration result between the first network output device and a target service provider node based on parsing results, computing resource status, and network resource status. The network management device is also used to distribute the first orchestration result to the network access device and send service request packets to the wide area network routing node, so that the service request packets are sent to the first network output device according to the specified path and strategy in the first orchestration result. The first network output device is used to send the service request packets to the target service provider node. By implementing the technical solution of this invention, more refined, flexible, and reliable end-to-end routing control is provided, achieving full-process protection of transmission and computing and optimal utilization of network and computing resources.

The present disclosure provides an FPGA hybrid wiring architecture with a spatially heterogeneous interconnection structure and an evaluation method. The FPGA hybrid wiring architecture with a spatially heterogeneous interconnection structure comprises: an FPGA chip divided into a plurality of sub-areas, each of the plurality of sub-areas being provided with a plurality of types of interconnection chips; each of the plurality of types of interconnection chips being provided with a wiring block connected with a logic block; the wiring block being provided to consist of one or more grouped areas, each area consisting of a first-level multiplexer group and a second-level multiplexer group, and the internal structure of the wiring block being determined according to a method of setting the internal structure of the wiring block. In addition, the present disclosure proposes a corresponding evaluation method for the FPGA hybrid wiring architecture with a spatially heterogeneous interconnection structure. The present disclosure aims to solve the problems of uneven resource utilization, serious wiring detour, and large area and delay waste of the traditional FPGA wiring structure.

Aspects of telemetry monitoring are described. Compute circuitry in a system includes hardware resources configured to perform compute operations. Telemetry management circuitry in the system is configured to detect a communication flow for an application function. The application function is associated with an application executing on at least one of the hardware resources. The telemetry management circuitry is configured to retrieve a telemetry configuration based on the communication flow. The telemetry configuration identifies at least one metric. The telemetry management circuitry is configured to execute a trace on the at least one metric to obtain metrics results. The telemetry management circuitry is configured to perform a telemetry action based on the metrics results.

Disclosed in the present invention are an IVI cross-application routing processing method, an electronic device and a routing architecture. The method comprises: on the basis of an annotation processor and ARouter logic, generating in a compilation stage a mapping relationship between a URL path and a class on the basis of annotation information marked in an application (S1); using a gradle plug-in to perform construction and management, obtaining routing information which is generated by the annotation processor and corresponds to the application, and writing the routing information into a configuration file (S2); during a startup process of a system arbitration application, dynamically scanning a configuration file of each application, so as to obtain routing information of each application, and generating a global routing table (S3); and initiating a routing jumping request by means of a corresponding URL path transmitted by the global routing table, and completing jumping (S4). In the present invention, the process of page jumping and navigation between applications is simplified, risks such as the complexity and error during a parameter transmission process are reduced, and an automobile application developer can be helped to construct a more efficient, maintainable and flexible application.

A cloud computing platform provides virtual machines on-demand to users. The platform provides a service that enables the user to access and control their virtual machine via an out-of-band channel for administrative purposes. A user invokes this service by authenticating with a control center and receiving a URL that points to a content delivery network (CDN). The user follows the URL to connect to a particular CDN server, per the CDN's request routing architecture. The CDN server applies instructions for handling this cloud compute platform URL, thereby decrypting an encrypted token in the URL to find, e.g., a network locator pointing to the host that is currently hosting the user's virtual machine, as well as an identifier for the user's virtual machine, and authenticators. The CDN server goes forward to the host, connecting to an endpoint which in turn connects to the user's virtual machine to provide the service.

This invention belongs to the field of constellation routing calculation technology, specifically disclosing a constellation network fusion method based on a multi-constellation interconnected distributed routing architecture. The entire routing architecture is divided into a satellite core autonomous system, a satellite access autonomous system, and a satellite user autonomous system. Gateways advertise terrestrial network routing prefix information to connected satellites, and network satellite terminals advertise the address prefixes of their connected user subnets to connected satellites. Satellites advertise routing prefixes to other satellites via internal boundary gateway protocols. When at least two satellites in the satellite core autonomous system are connected to different gateways within the same satellite access autonomous system on the ground, other satellites determine the next hop to reach the satellite access autonomous system based on the internal gateway protocol metric used to reach that satellite, improving the compatibility of satellite constellation network fusion at different construction stages.

Certain aspects of the disclosure are directed to routing data communications based on time zone. According to a specific example, a data-communications system for routing data communications based on time zone includes a communications routing circuit and a processing circuit. The communications routing circuit receives and routes data communications to a plurality of data communications stations and provides data communications services for users of the data communications stations. For such data communications routed in which a time zone routing feature is applied, the processing circuit determines a geographic location of the respective data communications station. The processing circuit then identifies a respective set of routing functions, based on a time zone of the geographic location, and routes the data communication based on the time zone, and according to the respective set of routing functions.

The invention discloses a routing architecture and a routing method based on AUTOSAR architecture cross-chip CAN data, and a vehicle. The routing architecture comprises a first chip and a second chip, a CAN interface module of the first chip receives CAN frames of a CAN bus and sends the CAN frames to a PDU routing module, the PDU routing module in the first chip directly routes the CAN frames to a conversion module in the first chip, and a protocol driving module in the first chip sends the CAN frames processed by the conversion module to the second chip through a preset protocol; and a Com module transplanted from the AUTOSAR in the second chip receives the packed CAN frame and unpacks the CAN frame to obtain a CAN signal. According to the invention, the problem that CPU resources are seriously consumed due to the fact that a CAN frame is routed by taking a signal as a unit and needs to be subjected to multi-layer processing of Com, Rte and SWC, hundreds of Rte interface calling is triggered by one frame of data in the prior art is solved, and the cross-chip CAN data transmission efficiency is improved.

Methods, systems and devices are disclosed including a substrate, a computational device mounted on the substrate, with a first surface of the substrate having one or more nanoelements formed within, the one or more nanoelements having a diameter of 100 nm-5,000 nm. In some embodiments, a heat dissipator may be mounted on the substrate, the heat dissipator having at least one nanostructure, and the one or more nanoelements forming a thermal conductive pathway between the computational device and the heat dissipator.

The application provides an on-site IO device on-demand communication routing system and method, relates to the technical field of device communication, and the system comprises an application arrangement layer, an on-demand routing decision layer and a device layer; the application arrangement layer outputs data to the on-demand routing decision layer; the on-demand routing decision layer issues configuration to the device layer; the device layer reports state data to the on-demand routing decision layer; the application arrangement layer is used for defining system control tasks and generating a standardized system model; the on-demand routing decision layer is used for intelligent decision of a communication mode and configuration conversion; the application arrangement layer, the on-demand routing decision layer and the device layer are logically linked in sequence and constitute an end-to-end on-site IO device on-demand communication routing architecture. The application realizes intelligent on-demand selection of a communication mode, takes into account the advantages of centralized scheduling and direct communication, improves the real-time performance of data interaction and the flexibility of the system, adapts to dynamic process requirements, and improves the intelligent level and the self-adaptive level of an industrial control system.

This invention provides a constellation network fusion method based on a tunnel-encapsulated multi-constellation interconnection routing architecture, belonging to the field of constellation routing calculation technology. The method divides the multi-constellation interconnection routing architecture into a satellite core autonomous system and a satellite user autonomous system according to the application scenario of the satellite network. The satellite core autonomous system adopts a distributed routing strategy and runs a customized internal gateway protocol. Border routers receive packets from the satellite user autonomous systems and encapsulate them with IP headers or link-layer headers before sending them to the satellites. The satellites receiving the packets forward the packets at the network layer or link layer to the destination network satellite terminal or destination gateway. The destination network satellite terminal or destination gateway decapsulates the packets, restoring them to the original IP packets, and sends the original IP packets to the destination network satellite terminal by looking up the routing table or forwarding table based on the destination IP address in the original packets, thus fusion of satellite constellation networks at different construction stages.

The invention provides an optimization method of digital currency cross-chain atomic exchange, and belongs to the technical field of digital currency. Historical transaction data acquisition and feature engineering processing are performed on a plurality of digital currency block chain systems; a CatBoost algorithm is adopted to establish each system prediction model, a transaction propagation diffusion mechanism is combined to construct an adaptive delay prediction algorithm, meanwhile, a layered deterministic wallet system is established to realize key security management, and a layered routing architecture and a multi-path concurrent processing mechanism are constructed to improve the throughput of the system. And path optimization decision is performed based on the double-layer game model to select the optimal transaction path, the overtime parameter of the Hash time lock is dynamically adjusted according to the prediction result, and the redundant path is switched to complete the atomic exchange process when the overtime parameter is abnormal, so that the technical problem of insufficient exchange time sequence control precision caused by network delay uncertainty in the cross-chain atomic exchange process is solved.

The present invention relates to technical fields related to resource scheduling optimization, specifically including a resource scheduling optimization method and system combined with cloud-edge collaboration, including: edge computing nodes configure resource requirements and traffic pressure based on interactive analysis of computing load, bandwidth fluctuations and cloud energy consumption centers, predict bandwidth trends and set traffic peaks and valleys; deploy lightweight routing architecture to initialize scheduling, perform bidirectional optimization with nodes as the center, adaptively balance loads, and obtain cloud-edge collaborative resource scheduling strategies. This solves the technical problem of insufficient accuracy in predicting network bandwidth change trends, inability to accurately capture the dynamic characteristics of traffic peaks and valleys, resulting in resource scheduling strategies lagging behind the actual network status, and further exacerbating transmission delays and bandwidth waste. This method establishes a dynamic prediction model based on node resource requirements and traffic pressure, accurately divides network traffic peak and valley intervals, and performs bidirectional resource configuration with edge nodes as the center in combination with traffic peak and valley characteristics, thereby improving bandwidth utilization.

This invention relates to an online-learnable neuromorphic processor based on approximate computation, comprising: a communication subsystem, a training subsystem, and an inference subsystem; the communication subsystem is used to construct a multi-pulse core neuromorphic processor with a hierarchical routing architecture; the training subsystem is used to optimize the weights and threshold parameters of the multi-pulse core neuromorphic processor using an online learning method; the inference subsystem is used for the multi-pulse core neuromorphic processor, after online training or offline weight deployment, to perform inference on encoded input image data and generate the classification output of the multi-pulse core neuromorphic processor. This invention proposes a neuromorphic processor architecture based on online learning and approximate computation to achieve a high degree of trade-off between accuracy, energy efficiency, and response speed, enabling real-time and efficient processing of dynamic and complex task data.

A system is provided which includes a first communications network including first nodes and a second communications network including a group of second nodes. Each of the first nodes is configured to transmit a multi-destination packet to two or more second nodes among the group of second nodes based on numerical identifiers which respectively correspond to the second nodes and are different from one another and a composite number corresponding to the group of second nodes. The composite number is a product of the numerical identifiers.

The application discloses an IVI cross-application route processing method, electronic equipment and a route architecture, and comprises the following steps: based on an annotation processor and an ARouter logic, generating a mapping relationship between a URL path and a class according to annotated annotation information in an application in a compilation stage; using a gradle plug-in to build and manage, obtaining route information corresponding to the application generated by the annotation processor, writing the route information into a configuration file, dynamically scanning configuration files of each application in a system arbitration application starting process, obtaining route information of each application program, generating a global route table, and initiating a route jump request and completing jump through a corresponding URL path delivered by the global route table. The application simplifies the process of inter-application page jump and navigation, reduces the complexity and error risk in the parameter transmission process, and can help automobile application developers to build more efficient, maintainable and flexible application programs.

The application discloses a lightweight routing design method and system suitable for a chain broadband wireless ad hoc network. The system comprises a configuration network node, a definition node routing architecture, a node-in configuration packet classifier module, a delay module, a general module, a neighbor list module, a neighbor information manager module, a delay estimator module, a packet queue module, a queue processing module, a MAC receiver module and a MAC transmitter module. The method is to classify and calculate the queuing of delay packet and general packet according to the data frame structure; to regularly update the neighbor list according to the data packet of the adjacent relationship; to estimate the time required for local queue transmission; to queue delay data and general data according to the priority; and to establish a forwarding link. The node of the application automatically completes path planning and queue sorting in real time according to the routing protocol architecture, does not need to be uniformly and centrally dispatched by the upper layer routing, and does not occupy network resources. The method maximally utilizes the node resources on the basis of guaranteeing the communication delay.

The invention relates to a nebula computing network technology of a dynamic low earth orbit satellite, in particular to a time delay sensitive task routing optimization method and system for a dynamic satellite network. The method comprises the following steps: deploying an inter-satellite routing architecture, defining a task parameter group, constructing a network communication model, establishing a time delay calculation model for task transmission processing, and determining a task dynamic routing strategy; according to a time delay calculation model and a task dynamic routing strategy, determining a target function and a constraint condition of an optimization problem by taking minimization of transmission delay and maximization of throughput as targets; and solving an objective function of the optimization problem based on reinforcement learning, and generating a dynamic inter-satellite routing strategy according to a solving result. According to the method, through fusion of spatial-temporal feature perception and reinforcement learning decision, the problem of unbalanced resource utilization under high dynamic topology is solved, the throughput of a satellite network is remarkably improved, the end-to-end delay is reduced, and efficient routing support is provided for an on-satellite computing task.