34 results about "Network cost" patented technology

The invention discloses a long-distance port expansion device based on a multi-port PHY chip, and the device comprises a near-end device which is integrated with a multi-channel serial interface supporting a multi-port protocol; the remote equipment comprises at least one PHY (Physical Layer) chip, and the PHY chip is provided with an uplink multi-channel protocol interface supporting a multi-port protocol and is provided with a plurality of downlink network ports; and the multi-channel serial interface of the near-end equipment is connected with the uplink multi-channel protocol interface of the PHY chip through a long-distance transmission medium. According to the invention, the port access and data forwarding functions of a traditional switch (including a switch realized by a switch chip or a processor) are directly replaced by the far-end multi-port PHY, and the method is suitable for various network communication scenes sensitive to network cost, time delay and architecture complexity.

The application provides a kind of evaluation method and system of neural network cost model based on category guidance, it is related to data processing technical field, the feature of the tensor program to be evaluated is extracted from the perspective of syntax tree, and the tensor program feature is obtained;Tensor program features are processed by dimension reduction and clustering, to obtain the probability that the tensor program feature belongs to different categories;The feature of the tensor program to be evaluated is extracted from the perspective of scheduling primitive, to obtain intermediate features;Based on the probability that the tensor program feature belongs to different categories and the intermediate features, the performance score of the tensor program to be evaluated is obtained.From the perspective of syntax tree, extracting the features of the tensor program and extracting the features from the scheduling primitive can effectively utilize the complex relationship between subprograms, reduce the time of the AI compiler automatic tuning process, so that the neural network model can be deployed on the ground, and only needs less tuning time under the same tuning performance condition, improves the efficiency.

Embodiments of the present application provide a method of obtaining a distribution key and related devices for obtaining a distribution key. The method is applied to cloud management platforms. This method includes: receiving a request from a tenant, where the request is used to obtain a set of target distribution keys; obtaining a set of target tables and a set of candidate distribution keys based on the request and a set of historical query information; obtaining a set of first distribution keys from the set of candidate distribution keys based on a first optimization rule; and obtaining the set of target distribution keys from the set of first distribution keys based on a network cost of each couple of first distribution keys. The proposed technique provides a suitable distribution key for tenant, improves the efficiency of obtaining a distribution key, and improves the efficiency of query.

The invention discloses a server-free edge function reliability deployment method and system based on reinforcement learning, and the method is oriented to server-free edge calculation, takes the minimization of long-term network cost as a target, and systematically solves the problems of high cold start cost, limited resources, difficult reliability guarantee and the like. A system model including user movement, channel fading, function deployment and a backup mechanism is constructed, four types of costs of cold start, maintenance, transmission and punishment are accurately described and converted into a Markov decision process, and a state-action-award framework is utilized to realize dynamic decision. The PPO algorithm fusing the greedy strategy to assist the action selection is further provided, and a function deployment scheme meeting the reliability constraint is efficiently generated under the condition that resources are limited. According to the method, through fusion of reinforcement learning and a greedy strategy, the problems of feasibility, reliability and cost of function deployment in server-free edge computing are solved, and the method has the capabilities of efficient deployment, adaptive scheduling and long-term performance optimization.

The application is suitable for a large-bandwidth network flow scheduling management method and system, the method comprising: acquiring flow restriction parameters and determining maximum network bandwidth; monitoring the demand of each network port, sampling the bandwidth demand of each network port according to a preset sampling demand monitoring frequency, and generating bandwidth demand coordinates; constructing a network port bandwidth demand prediction function corresponding to each network port, synthesizing a comprehensive demand prediction function based on the network port bandwidth demand prediction function; predicting the bandwidth within a preset time length according to the comprehensive demand prediction function, obtaining prediction bandwidth data, and allocating network bandwidth to each network port. The application determines the sampling data of each network port by high-frequency sampling, constructs a prediction function of each network port based on the sampling data, predicts the bandwidth change within a short time according to the prediction function, and schedules the bandwidth of the network port according to the prediction, so as to fully utilize the 95 billing rules and save network cost.

The invention discloses a data routing system. The data routing system comprises a first topology network and a second topology network. The first topology network is nested in the second topology network. The network structure of the first topology network is a fat tree network topology structure formed by a first node group and a switch, the first node group comprises a first number of network nodes, and each network node in the first topology network is in cascade connection with the switch. The network structure of the second topology network is a full interconnection network topology structure formed by second node groups, the second node groups comprise a second number of first node groups, and the second number of first node groups are connected. According to the system, a multi-stage hierarchical network structure with a nested fat tree network topology and a full-interconnection network topology is established through small switching nodes with a small number of ports, so that a large-scale or super-large-scale network is established, the network cost and power consumption are effectively reduced, and the network expandability is improved.

The present application provides a network slice orchestration method for a programmable data plane network. After obtaining service requests corresponding to different network slices to be orchestrated, the method uses a configured network cost function, a sequence of VNF types to be traversed corresponding to any service request, and VNF instances of the corresponding VNF type, the required maximum routing delay, and the source and destination nodes of each network slice to be orchestrated to perform node deployment on the VNF, thereby obtaining the optimal routing path and VNF deployment information corresponding to the service request. Based on the optimal routing path and VNF deployment information corresponding to the service request, a corresponding orchestration file is generated, and the orchestration file is distributed to each switch in the programmable data plane network to orchestrate the network slice corresponding to the service request. This method enables the coordinated optimization of resource allocation, path selection, and service quality assurance, thereby improving the flexibility and efficiency of network slice deployment and resource allocation.

A computing system generates directives of a program. Beam nodes are selected one level at a time from multiple nodes in the tree structure. Each node represents a subset of operations in the program. A first number of the beam nodes are selected at a given level of the tree structure. The selection of the first number of the beam nodes uses a cost model that is based on a neural network. A second number of the beam nodes are selected using a random search. The ratio of the first number to the second number is determined based on a search completion percentage at the given level. A path is identified that passes through respective beam nodes at multiple levels of the tree structure. The path represents a schedule for executing the program on a target machine. Then the directives corresponding to the schedule are generated.

The application discloses a kind of intelligent routing method, equipment and storage medium based on link quality, by real-time measurement and monitoring to tunnel quality on equipment, including link delay, bandwidth utilization, packet loss rate and other performance indicators, and upload to SDWAN control platform, provide basic data for intelligent routing.Based on these data, after calculation, make SD-WAN according to real-time network and application, select the optimal path to carry out data transmission.By the method of the application, customers can flexibly configure link quality threshold of different application groups according to needs, including packet loss (%), delay (ms), jitter (ms), when network link performance declines, specific application traffic can be identified, and the intelligent routing scheme of selecting the optimal path for application group to transmit data traffic.The application simplifies user configuration, improves network quality, and effectively reduces user network cost.

The invention discloses a main network construction cost rapid auxiliary review method and system, and the method comprises the steps: firstly obtaining construction cost data information and construction cost budget information of a main network construction cost, and recognizing each construction cost demand type and construction cost data distribution information based on the construction cost data information; then, based on the cost budget information, identifying a cost list, and based on the cost data distribution information and the cost list, identifying cost evaluation index values and cost evaluation reasons; then, based on the cost evaluation index value and the cost evaluation reason, identifying a cost evaluation result, and based on the cost evaluation result, generating main network auxiliary evaluation information of the main network cost; according to the method, the cost evaluation result of each cost demand type is comprehensively analyzed, workers are assisted to comprehensively analyze each cost demand type, and each cost demand type is comprehensively analyzed through the evaluation index, so that the timeliness, the accuracy and the comprehensiveness of main network cost evaluation analysis are improved, and the efficiency of main network cost evaluation analysis is improved. And meanwhile, the review efficiency is further improved.

The invention relates to the technical field of fifth-generation mobile communication core networks, in particular to a method and a device for realizing fixed wireless access security enhancement in a 5G network. The method is executed by a 5G core network, and comprises the following steps: in a PPPoE session discovery stage of accessing an FWA terminal to an L2 transmission network, acquiring terminal equipment security information, generating security enhancement information based on a pre-configured security policy, carrying the information through a Relay-Session-Id tag of a PPPoE protocol standard, and sending the information to a PPPoE server, thereby realizing dual authentication. The scheme adapts to IP type and Ethernet type PDU session scenes, does not need to additionally deploy hardware, is compatible with multiple types of terminals, improves the access security, reduces the network cost, and is suitable for multiple types of access scenes such as families, enterprises, remote areas and the like.

The invention discloses a batch high-threshold active secret sharing method and device, and the method comprises the steps: building a high-threshold active secret sharing basis through a high-threshold module, adding a blind value for a distributor through the high-threshold module, and enabling the blind value to be shared with a secret s; realizing secret sharing of a reconstruction threshold between t + 1 and n-t in combination with a high-threshold asynchronous complete secret sharing protocol and a multi-valued verification Byzantine negotiation protocol; based on a high-threshold module, a dual-committee asynchronous complete secret sharing protocol and a randomness extraction technology based on an ultra-reversible matrix are adopted to reduce the protocol complexity and realize batch secret transfer through a batch sales sharing module; a batch high-threshold active secret sharing protocol Bat-DPSS constructed by using a high-threshold module and a batch sharing module adapts to an asynchronous or partially synchronous network environment without depending on a public key infrastructure PKI, and can support dynamic replacement of committee members and resist attacks of mobile opponents. According to the method, the network cost is low after distribution, and the efficiency is improved compared with that of an asynchronous DPSS protocol.

The invention provides an evaluation method and system for a neural network cost model based on category guidance, and relates to the technical field of data processing, and the method comprises the steps: extracting the features of a to-be-evaluated tensor program from the perspective of a syntax tree, and obtaining the features of the tensor program; performing dimension reduction processing and clustering processing on the tensor program features to obtain probabilities that the tensor program features belong to different categories; extracting features of the tensor program to be evaluated from the perspective of scheduling primitives to obtain intermediate features; and obtaining a performance score of the to-be-evaluated tensor program based on the probabilities that the tensor program features belong to different categories and the intermediate features. The tensor program features are extracted from the syntax tree perspective and the features are extracted from the scheduling primitive perspective, so that the complex relation between subprograms can be effectively utilized, the time of the automatic tuning process of an AI compiler is shortened, and when a neural network model is deployed in a landing mode, only less tuning time is needed under the condition that the same performance after tuning is achieved, and the efficiency is improved.

The invention provides a traffic processing method and device, electronic equipment, a computer program product and a computer readable storage medium. The method comprises the following steps: acquiring a plurality of flow data sets collected from a network system, and performing feature extraction on each flow data set to obtain feature data corresponding to each flow data set; performing clustering processing on the plurality of flow data sets based on the feature data corresponding to each flow data set to obtain a clustering result; determining a feature tag of each traffic data set based on a clustering result; and scheduling each traffic data set based on the feature tag of each traffic data set. According to the method and the device, the traffic data can be flexibly scheduled, the network performance is improved, and the network cost is reduced.

The invention relates to the technical field of mobile positioning, in particular to a terminal positioning method and device, computer equipment and a readable storage medium. The method comprises the following steps: receiving a first new air interface positioning assistance protocol NRPPa message sent by a location management function LMF network element through a base station; sending a first NRPPa message to the base station, so that the base station controls a TRP connected with the base station based on the first NRPPa message, and sending a scanning indication message to the Bluetooth scanning device through the peripheral interface; further, the base station obtains a Bluetooth AOA scanning result, and sends a second NRPPa message to the LMF network element; wherein the second NRPPa message carries a Bluetooth AOA scanning result, and the second NRPPa message is used for triggering the LMF network element to carry out terminal positioning based on the Bluetooth AOA scanning result and an uplink signal time difference of arrival (UTDOA) measurement result; according to the invention, the Bluetooth scanning equipment is effectively simplified, and the gateway cost and the network cost required by terminal positioning are reduced.

The invention discloses a data processing method and device, equipment and a readable storage medium, and the method comprises the steps: obtaining a first network connection state of first terminal equipment, and if a network connected with the first terminal equipment belongs to a local area network type and service information of second terminal equipment is detected in the network, determining that the first terminal equipment is in a first network connection state; if so, forwarding streaming verification information sent by the second terminal device to the first terminal device, and obtaining a verification result; if the network indicated by the first network connection state belongs to a public network type, or service information of the second terminal equipment is not detected in the network indicated by the first network connection state, obtaining a verification result from a background server, and establishing an indirect connection relationship for performing streaming service between the first terminal device and the second terminal device through the verification result. By adopting the streaming service method and device, the streaming service is carried out between the two terminal devices of any network type, the compatibility is improved, a background server does not need to carry out data forwarding in the streaming, and the network cost is reduced.

The invention discloses a network utility optimization method for a multi-unmanned aerial vehicle cooperative mobile edge computing network, and belongs to the field of wireless communication. The implementation method of the invention comprises the following steps: constructing a multi-UAV collaborative MEC network architecture; respectively constructing a communication and calculation model, a network cost model and a network utility model by combining the unloading selection variable and the calculation selection variable; 2, constructing an optimization problem by taking maximization of the network utility model obtained in the step 1 as a target; the UPTSM algorithm decomposes an original optimization problem into two sub-problems, and the two sub-problems are converted into two bipartite graph matching problems; sequentially solving the two problems by adopting a Hungary algorithm to obtain an unloading selection variable and a calculation selection variable; the VC-WOA limits variables violating the task processing time limit to quickly search legal solutions, so that the network utility is improved to the greatest extent. According to the invention, by optimizing multi-UAV cooperative task scheduling and energy management, the cooperation efficiency can be improved, and fair task distribution is realized, so that the network income is maximized.

Disclosed in the present application is a data routing system. The system comprises a first topological network and a second topological network, wherein the first topological network is nested in the second topological network; the network structure of the first topological network is a fat-tree network topology structure consisting of a first node group and a switch, the first node group comprises a first number of network nodes, and there is a cascade relationship between the network nodes and the switch in the first topological network; and the network structure of the second topological network is a full-mesh network topology structure consisting of a second node group, the second node group comprises a second number of first node groups, and there is a connection relationship among the second number of first node groups. In the system, a multilevel hierarchical network structure where a fat-tree network topology is nested in a full-mesh network topology is established by means of small switching nodes having a small number of ports, and thus a large-scale or ultra-large-scale network is established, thereby effectively reducing the network cost and power consumption and improving the network scalability.

A method (300) for controlling an energy storage device or load comprises generating a hybrid tariff based on a weighted combination of a tariff and network cost. An energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction are generated (302,304) for time periods for each site in a plurality of sites. The method further comprises generating (306), for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction.

Embodiments of the present disclosure provide method and apparatus for cluster management. A method performed by a first node of a cluster comprises receiving equal network cost group (ENCG) data for the cluster from a cluster management node. The method may further comprise generating a first network cost matrix for the cluster based on the ENCG data for the cluster.

Systems and methods for application performance management across one or more networks are disclosed. A system includes a plurality of geographically distributed computing devices executing one or more applications. A plurality of collectors are distributed across the one or more networks, each collector being positioned proximate a respective computing device. The collectors may sample, by each of the plurality of collectors, a plurality of trace events received from the applications executing on the computing devices proximate the collector. The collectors may retain in memory sampled trace events that meet a configurable attribute. The collectors may use probabilistic sets and tail-based sampling to determine root events and policies for identifying relevant traces. Benefits can be achieved in bandwidth savings, network cost and cyber security.

A system and method (300) for controlling an energy storage device or load are provided. The method comprises generating a hybrid tariff based on a weighted combination of a tariff and network cost. An energy usage prediction and, where renewable or other generation sources are available, an energy generation prediction are generated (302, 304) for time periods for each site in a plurality of sites. The method further comprises generating (306), for the site, a plan to selectively store energy in an energy storage device at the site, discharge energy from the energy storage device, or control when a load at the given site consumes energy based at least in part on optimising the cost of energy consumed according to the hybrid tariff based on the energy usage and generation prediction. [Figure 3]

The present disclosure relates to a traffic control method, device, equipment and program product. The traffic control method applied to a client gateway module comprises: determining a target service in response to a data request sent by a client service module; determining a target time window corresponding to the target service; wherein the target time window is determined based on a target traffic model corresponding to the target service, and the target traffic model is used to balance network traffic; and sending the data request to the server gateway module within the target time window to request target data of the target service from a server service module corresponding to the server gateway module. In this way, the data requests of multiple clients can be more evenly dispersed, the network traffic can be better flattened, and traffic peaks and server peak capacity can be avoided, thereby reducing the network cost of data resource requests and the resource consumption of the server.

The invention relates to the field of network resource allocation, in particular to an FSDM network limited uplink and downlink port dynamic survival optical path configuration method and system, and the method comprises the steps: constructing an FSDM optical network employing a shared laser source and a local oscillator, building an analytical model, and predicting the network blocking performance under the condition of different numbers of uplink and downlink ports of optical cross connection nodes, determining a port threshold; and on the basis of a port threshold value, whether an idle uplink and downlink port exists in a source node and a destination node or not is firstly judged through a constructed adaptive routing algorithm, and then working and protection path search is started, so that FSDM optical network dynamic survival optical path configuration based on shared backup path protection is realized. According to the FSDM optical network protection strategy, the blank of the FSDM optical network protection strategy is filled, the analytical model can reduce redundant configuration and optimize the network cost, the adaptive routing algorithm can improve the spectrum utilization efficiency, and the medium Spa SCh granularity configuration is adopted, so that the FSDM network has high cost benefit.

The invention discloses a pipe network optimization method and system for a gas-steam combined cycle cogeneration unit, and relates to the field of gas-steam combined cycle cogeneration units. Aiming at the problem that the comprehensive cost of a unit and a steam pipe network is difficult to optimize in the traditional step-by-step design, the method comprises the following steps: constructing a plurality of steam pipe network schemes with different pipe section diameters and determining the total annual investment cost; determining unit steam supply parameters according to the pipe network inlet steam parameters; establishing a bottom-layer steam circulation thermal model, and determining unit heat consumption based on steam supply parameters; and calculating the annual comprehensive cost by combining the annual total investment cost and the unit annual operation income, and finally screening out the optimal steam pipe network scheme with the lowest annual comprehensive cost. The system correspondingly comprises a pipe network cost module, a steam supply parameter module, a heat consumption module and a pipe network optimization module. According to the invention, collaborative optimization of the steam pipe network and the unit is realized, the comprehensive cost is effectively reduced, and the system economy is improved.

The application discloses a data stream processing method and device between different task nodes and electronic equipment, and relates to the technical field of big data. The method comprises the following steps: performing a modulo operation on a hash value corresponding to an extreme value in a first data stream sent by an upstream task node, and dividing the first data stream into L data blocks, wherein the cut point positions corresponding to the L data blocks are determined by the result of the modulo operation; when a modification operation occurs in the first data stream, recording the modification time and modification content of the first data stream, wherein the modification content comprises the change of the data blocks in the first data stream; and controlling the second data stream update of a downstream task node according to the modification time and modification content of the first data stream, wherein the second data stream is the first data stream before the modification that is saved by the downstream task node. The application solves the technical problem of high transmission network cost in the prior art when data streams are transmitted between different task nodes.

The application relates to an Internet of Things edge computing P2P data collection method and system, which comprises the following steps: an edge gateway is connected to an Internet of Things platform through an MQTTS protocol; an Internet of Things device integrates an Internet of Things SDK, starts the SDK and is connected to the Internet of Things platform, and the connection mode is selected according to whether the edge gateway has a public network IP; the Internet of Things device communicates with the SDK, the SDK pushes messages to the edge gateway according to the connection mode, and the edge gateway uploads the Internet of Things device connection information to the Internet of Things platform; the edge gateway and the Internet of Things platform realize message uploading and issuing through the MQTTS; a long connection disconnection duration threshold value is set, when the long connection disconnection duration exceeds the threshold value, it can be judged that the edge gateway is down, and the Internet of Things device is re-distributed. The Internet of Things device is connected to the Internet of Things platform by integrating the SDK, the massive data is transmitted to the Internet of Things platform after being calculated by the edge gateway, the hardware and network costs of the Internet of Things platform are reduced, the Internet of Things platform can be connected regardless of whether the edge gateway has a public network IP, and the computing power is provided for the Internet of Things device.

Disclosed in the present application is a data routing system. The system comprises a first topological network and a second topological network, wherein the first topological network is nested in the second topological network; the network structure of the first topological network is a fat-tree network topology structure consisting of a first node group and a switch, the first node group comprises a first number of network nodes, and there is a cascade relationship between the network nodes and the switch in the first topological network; and the network structure of the second topological network is a full-mesh network topology structure consisting of a second node group, the second node group comprises a second number of first node groups, and there is a connection relationship among the second number of first node groups. In the system, a multilevel hierarchical network structure where a fat-tree network topology is nested in a full-mesh network topology is established by means of small switching nodes having a small number of ports, and thus a large-scale or ultra-large-scale network is established, thereby effectively reducing the network cost and power consumption and improving the network scalability.

The application provides a random access method, system and related device. The method comprises: receiving a random access request sent by a mobile terminal in a first cell; allocating a user identifier according to the random access request; and sending a first message containing the user identifier and a signaling request to a corresponding second cell, the first message being used to assist the second cell to determine the user identifier and to establish a signaling bearer with the mobile terminal; wherein the first cell is a cell processing data plane functions, a first base station is a base station where the first cell is located, the second cell is a cell processing signaling plane functions, and a second base station is a base station where the second cell is located. The application uses random access resources of the first cell to perform a random access process and to establish a signaling bearer between the second cell and the mobile terminal, so as to reduce network cost and energy consumption.