255 results about "Goal node" patented technology

The application provides a GPU resource intelligent scheduling method and system based on reinforcement learning, relates to the field of cloud computing resource scheduling, and comprises the following steps: collecting resource state data of a GPU cluster and attribute data of a to-be-scheduled task, pre-processing, constructing a reinforcement learning model, and training the reinforcement learning model by using historical scheduling records; inputting the resource state data of the current GPU cluster and the task attribute data of the to-be-scheduled task into the trained reinforcement learning model to obtain a preliminary scheduling scheme; generating a candidate scheduling scheme set based on the preliminary scheduling scheme, selecting a final scheduling scheme from the candidate scheduling scheme set by using a multi-objective optimization algorithm; performing resource slice configuration on a target GPU node in the GPU cluster according to the final scheduling scheme, and starting the to-be-scheduled task in a task execution container; and updating the parameters of the reinforcement learning model by using experience data. The application realizes adaptive evolution and multi-objective balanced optimization of a scheduling strategy.

This application discloses a path planning method, apparatus, medium, and product for an aircraft traction system in confined spaces, relating to the field of aircraft path planning technology. The method includes: establishing a kinematic model of the aircraft traction system located in a hangar; constructing a path planning mathematical model based on the kinematic model; employing an improved hybrid algorithm to perform path search on the aircraft traction system based on the path planning mathematical model to obtain the target path of the aircraft traction system; the improved hybrid algorithm is derived from a hybrid algorithm, and the improvements include: using RS curves for bidirectional node expansion; setting the cost function to include the cumulative historical cost from the starting node to the current node and the heuristic cost from the current node to the target node; and establishing a closed table distinguishing between forward and backward directions. This application realizes path planning for an aircraft traction system in confined spaces.

This application discloses a method, apparatus, electronic device, and storage medium for splitting a storage cluster, relating to the field of computer technology, particularly to artificial intelligence fields such as cloud computing, distributed storage, and big data processing. The specific implementation scheme is as follows: The data path of the target resource in the original storage cluster is adjusted so that requests to access the target resource are sent to the target node in the original storage cluster; the target node is network isolated; the target node is restarted, and in response to the target node's target startup parameter being set to the target value, the target node is split from the original storage cluster as a new storage cluster; wherein, the target startup parameter is a parameter used to force a node to start as a new single-node cluster; the target node is removed from the original storage cluster, and the network isolation of the target node is lifted.

The application relates to the technical field of intelligent operation and maintenance of industrial equipment, and in particular to a large model industrial knowledge retrieval enhancement method and system and a storage medium. Text blocks of an industrial knowledge text set are mapped into text feature vectors, and are stored in each node of a fault tree according to semantic matching relationships to construct a topological vector database; real-time sensor data of equipment are collected, and a deviation matrix is calculated to lock an initial target node in the fault tree; after a user query request is vectorized, similarity retrieval is performed in a local vector subset of the target node, and when the retrieval score does not satisfy a confidence threshold, the retrieval range is expanded back to the parent node along the fault tree; and text blocks satisfying the condition are jointly input into a large language model together with the query request to generate a diagnostic reply. The application can solve or at least alleviate the problems of false recall and large model diagnosis illusion caused by semantic overlap of industrial vocabularies in existing global vector retrieval, and provides a large model industrial knowledge retrieval enhancement method, system and storage medium.

This application provides a scheduling method, which includes: a scheduling system receiving a resource request or a service request; wherein the resource request or service request includes first resource information or first service information and demand information; wherein the demand information includes at least performance demand information and energy consumption demand information; responding to the resource request or service request, and determining a target node based on the first resource information or first service information, demand information, and node information; wherein the node information includes node performance information and node energy consumption information. This application differs from the traditional nearest-neighbor scheduling method by comprehensively considering the node's performance information and energy consumption information, thereby scheduling service requests to the optimal target node and improving the computational efficiency of network resources.

The application discloses a kind of multi-node server network topology and energy consumption perception's task scheduling method, belong to data center and high performance computing technical field.It includes: constructing the weighted directed graph topology model including server node and switch node, the dynamic energy consumption weight function of link is defined;Establish the three-dimensional energy consumption and thermal interference evaluation model of node, introduce space thermal interference factor;Analysis the characteristics of task to be scheduled;Based on energy consumption potential field screening candidate node;Build the global energy consumption cost function including computing energy consumption, network transmission energy consumption and thermal coupling penalty energy consumption;According to the principle of minimum total energy consumption, select target node to execute scheduling and update topology state.The application solves the problem of splitting computing and network energy consumption, ignoring the space coupling effect of heat dissipation in the prior art by the fusion of network topology perception and multi-dimensional energy consumption model, and realizes the optimization of global energy consumption of data center.

According to an embodiment of the present disclosure, provided are a method and electronic device for training a target estimation artificial intelligence (AI) model. The method may comprise the steps of: identifying a target node corresponding to a target estimation AI model; identifying a node related to the target node on the basis of edge weights and edge types in a knowledge graph; selecting a feature of data on the basis of the node related to the target node; and training the target estimation AI model so as to output target information by using selected data including the selected feature.

The application relates to a kind of OFDM-based broadband millimeter wave relay system hybrid beamforming design method, comprising: calculating the receiving signal of target node under the condition that each node is all digital processor;The receiving signal is processed using minimum mean square error criterion in target node, and MMSE matrix is calculated;Using the equivalence between maximizing sum rate and minimum weighted mean square error algorithm, the maximum sum rate problem is converted into minimum weighted mean square error problem according to MMSE matrix;Deep unfolding neural network is used to solve the minimum weighted mean square error problem, and all digital processor of each node is obtained;Each node's all digital processor is decomposed based on least square decomposition algorithm, and hybrid beamforming matrix of each node is calculated, the complexity of the algorithm is greatly reduced, and the running time of the system can be effectively improved.

This disclosure provides a node anomaly detection method, apparatus, electronic device, and storage medium. The method is applied to a computing cluster comprising multiple computing nodes. The method includes: performing risk prediction on a target node to be predicted among the multiple computing nodes; determining the predicted anomaly probability of the target node for multiple anomaly types, including multiple types such as faults, slow nodes, and silent data corruption (SDC); identifying risk nodes from the target nodes based on the predicted anomaly probabilities; performing anomaly detection on the risk nodes based on the predicted anomaly probability of each risk node to obtain anomaly detection results for the risk nodes; and identifying anomaly nodes from the risk nodes based on the anomaly detection results. Embodiments of this disclosure can improve the overall performance of the computing cluster and the proportion of computing nodes in the cluster that are in an available state.

The application provides a trust evaluation method of a multi-layer relationship-aware graph neural network, belongs to the technical field of social network data processing and graph learning, and aims to solve the problems of mixed multi-source trust information, insufficient modeling of nonlinear interaction between nodes, and insufficient utilization of multi-layer graph convolution features of existing trust evaluation methods in sparse social networks. The method constructs a directed trust graph, uses a relationship-aware encoder to model the node's own features and neighborhood features, combines an adaptive propagation module to represent the directional and conditional dependence relationship between nodes, and obtains the final node pair representation through cross-layer attention fusion to realize trust evaluation between target node pairs. The method can be applied to social platform risk control, recommendation system credibility analysis, transaction network reputation evaluation and other scenes.

This invention provides a method, medium, product, and device for session connection management during database playback. The method includes: receiving a session identifier from a session during playback; generating an allocation identifier based on the low-order byte of the session identifier; determining a target pointer array corresponding to the session in a pre-set session management table based on the allocation identifier, wherein the session management table pre-sets a predetermined number of pointer arrays, each pointer array corresponding to a singly linked list and pointing to the head node of the corresponding singly linked list; determining whether a target node matching the session identifier exists in the target singly linked list corresponding to the target pointer array; if no target node exists in the target singly linked list, creating a new connection node and inserting the new connection node into the target singly linked list. This method enables rapid generation of the allocation identifier based on the low-order byte, and achieves rapid location and efficient storage of session nodes using a singly linked list structure, ensuring the stability and smoothness of high-concurrency playback processes.

The application discloses a node classification method and device based on a decoupling graph, equipment, medium and product, and relates to the technical field of neural networks. The method comprises the following steps: obtaining a text attribute graph and an input prompt; inputting text information corresponding to each node into a large language model to obtain overall text representation corresponding to each node; inputting overall text representation corresponding to a target node into multiple decoupling channels, each decoupling channel aggregates neighbor information of the target node according to different graph structures to obtain multiple hidden layer representations; optimizing the input prompt based on the multiple hidden layer representations of the target node to obtain an optimized prompt, and inputting the optimized prompt into the large language model to obtain a node classification result. The application obtains hidden layer features containing different aspect graph structure information through decoupling channels, and optimizes the input prompt based on the hidden layer features, so that the large language model can better handle the node classification task of the text attribute graph, and effectively improves the processing performance of the large language model on the text attribute graph.

The application provides a resource scheduling method and device, a computer program product and a computer readable storage medium. The method comprises: obtaining a to-be-scheduled workload by a scheduler; determining a plurality of candidate nodes based on a first resource requirement of each resource type of the to-be-scheduled workload, a first allocable resource amount of each resource type of each node in a node cluster, and a total resource load on each node; determining, for each resource type, an average predicted resource usage of the resource type in the node cluster if the to-be-scheduled workload is scheduled to each candidate node, and determining a target resource type of the to-be-scheduled workload based on the average predicted resource usage; determining a load balancing index if the to-be-scheduled workload is scheduled to each candidate node based on the target resource type, and scoring each candidate node based on the load balancing index; determining a target node from the plurality of candidate nodes based on the scoring result, and scheduling the to-be-scheduled workload to the target node.

The embodiment of the present application provides a method and device for installing components, which are used for a cloud computing system, and the method comprises the following steps: receiving parameter information, wherein the parameter information at least comprises component information of a target component to be installed; determining a target node for installing the target component in each node of the cloud computing system according to the parameter information, and establishing a connection with the target node; searching for an installation package corresponding to the target component in a preset installation source according to current version information of the cloud computing system and the component information of the target component; and transmitting the installation package to the target node, so that the target node realizes automatic installation of the target component by running the installation package. Through the present application, the target component can be automatically installed under the cloud computing system, so that the labor cost can be reduced and the efficiency of installing components can be improved.

This invention discloses a routing method and system for a multi-layered, multi-path switching network. The method includes the following steps: constructing a directed graph based on the multi-layered network, wherein nodes in the directed graph represent switches, edges represent communication links, each node has the ability to switch between different switching technologies, and the communication link has the ability to transmit data according to the switching technologies of its two endpoints; for each node in the directed graph, initializing the routing table of that node and iteratively calculating the optimal routing table; given a starting node and a target node, obtaining the next hop of the minimum-cost continuous path from the starting node to the target node according to the optimal routing table of the current node, and sending data packets hop-by-hop to the next node. The routing algorithm proposed in this invention can solve the minimum-cost path problem between all node pairs.

The application relates to a job scheduling system, equipment and medium for strong and weak ability mixed nodes, the system comprising: a coordinator module configured to maintain a cluster topology structure and execute a consistency election protocol; a scheduler module configured to generate a task dispatch strategy according to a global resource pool state and a job dependency relationship; an executor module configured to receive and execute a job instruction issued by the scheduler module; a trigger module configured to listen to an external event source and convert an event meeting a preset condition into a job scheduling request and send the job scheduling request to the scheduler module; and a communication abstraction layer configured to shield transmission differences of a bottom layer from an upper layer application according to attributes of a target node and a uniform resource identifier prefix. In the above embodiment of the application, the election protocol integrated with the node capability mask field completely solves the problem that a weak node is mistakenly selected as a master node, leading to instability of a cluster. The application effectively adapts to a narrowband and high-delay network environment and improves communication reliability in a weak network.

The application provides a resource scheduling method and an electronic device, which can be applied to the technical field of computers. The resource scheduling method comprises the following steps: in response to a resource request for a target task, obtaining resource utilization scores of nodes and hardware accelerators in a cluster according to resource usage amounts of the hardware accelerators, resource allocation amounts of the nodes, and a resource demand amount carried by the resource request; determining a target node in the nodes and a target hardware accelerator of the target node according to the resource utilization scores and a scheduling strategy carried by the resource request, the scheduling strategy indicating a scheduling rule matched with resource integration requirements and / or load balancing requirements of the target task; virtually dividing the target hardware accelerator according to the resource demand amount to obtain a virtual hardware accelerator; and binding the virtual hardware accelerator to a container of the target node to execute the target task by using the container.

This application relates to the field of power system technology and discloses a branching method and apparatus for accelerating the solution of power system safety-constrained unit combinations. During the branching process, if the relaxation problem of the target node is unsolvable, instead of simple single-point pruning, a network constraint minimum violation model is constructed. If the network constraint results in no solution, a globally feasible cut is generated using the Lagrange multiplier at the optimal solution. This globally feasible cut represents the constraint condition of a specific unit combination mode that leads to the unsolvability of the relaxation problem. Adding the globally feasible cut to the branching cut pool allows for the removal of infeasible solutions for similar specific unit combination modes in the future, avoiding a large number of repetitive and invalid explorations during the branching process, thereby accelerating the solution and ensuring the safety and economy of power grid dispatch. In this application, the globally feasible cut is strictly based on strong duality theory, ensuring that feasible or optimal solutions are not erroneously removed, fundamentally guaranteeing the feasibility and optimality of the solution results for safety-constrained unit combinations.

This application relates to the field of data analysis technology, and in particular to a method, apparatus, computer device, storage medium, and computer program product for extracting profile lines from oblique photogrammetry data. The method includes: dividing the oblique photogrammetry data into multiple block files according to a geographic grid, with each block file containing a root node corresponding to a specific block file, and hierarchically storing node files corresponding to child nodes derived from the root node; identifying target block files and target node files that intersect with the projection line data; reading specific geometric data from the target node files, with the geometric data associated with profile line extraction; using a preset intersection algorithm to traverse the geometric data in all target node files to obtain profile points, and connecting the profile points to generate target profile lines. This method can improve the efficiency of profile line extraction in large-scale oblique photogrammetry data and reduce the computational load of profile line extraction.

Computer systems and computer-implemented methods modify a machine learning network, such as a deep neural network, to introduce judgment to the network. A “combining” node is added to the network, to thereby generate a modified network, where activation of the combining node is based, at least in part, on output from a subject node of the network. The computer system then trains the modified network by, for each training data item in a set of training data, performing forward and back propagation computations through the modified network, where the backward propagation computation through the modified network comprises computing estimated partial derivatives of an error function of an objective for the network, except that the combining node selectively blocks back-propagation of estimated partial derivatives to the subject node, even though activation of the combining node is based on the activation of the subject node.

This specification provides a problem-solving method and apparatus, a computing device, and a computer-readable storage medium. The method includes: retrieving related nodes of a target problem from a semantic index graph; decomposing the target problem into multiple atomic search instructions based on the target problem, the graph structure of the semantic index graph, the related nodes, and a preset maintenance experience template; retrieving the target node corresponding to each atomic search instruction from the semantic index graph; and determining the target answer based on the target node corresponding to each atomic search instruction and the target problem. This method, through problem decomposition and graph retrieval based on semantic index graphs and expert experience, effectively improves the accuracy, completeness, and operability of solving complex problems in vertical domains.

This invention relates to the field of data processing technology, specifically to a method for dynamic positioning of page elements based on an adaptive waiting strategy. The method involves extracting latency jitter data and combining it with the historical loading characteristics of the target page to obtain a waiting time limit. Within this limit, the method monitors the internal mounting state, visual visibility state, and layout stability state of the target node. When the internal mounting state of the target node is ready, but the visual visibility state and layout stability state assessments fail, the geometric size feature weights of the target node are stripped, and node data of the internal mounting state is extracted. An undirected graph data structure is constructed to downgrade and anchor the target node. When the target node is downgraded and its local node block is already mounted to the page node tree, the nearest ancestor node container that has completed layout rendering is backtracked as a reference to construct a safe interaction boundary. Before the overall webpage resources are fully loaded, interactive operations are performed on target nodes that have not escaped the safe interaction boundary.

The application discloses a node routing method and device based on a behavior chain, equipment, a medium and a program product. The method comprises the following steps: receiving a user request; loading an initial behavior chain corresponding to the user request; the initial behavior chain is composed of an agent node and a routing node; when a first node in the initial behavior chain is a first agent node, calling the first agent to output a first output result; and recording the first output result to a global context state; continuing to execute the initial behavior chain, and when a second node in the initial behavior chain is a routing node, determining a target node based on the global context state; when the target node is not a node in the initial behavior chain, loading a target behavior chain corresponding to the target node, and continuing to execute the target behavior chain. The method can solve the problems of poor system expansibility, fragmented interaction and insufficient continuity.

This application discloses a resource scheduling method and related apparatus. The method includes: obtaining a first resource utilization rate for each node; wherein, a node is used to provide resources to a container group according to a scheduling request from a container group deployed on the node; filtering each node according to a first utilization rate threshold to obtain at least one target node; wherein, the first resource utilization rate of the target node is less than or equal to the first utilization rate threshold; for each target node, applying a scoring mechanism that matches the priority of the target container group, and scoring the target node according to its second resource utilization rate and resource allocation rate; and implementing resource scheduling according to the priority of the target container group and the scores of each target node corresponding to the priority. This achieves load balancing among nodes and improves the operational stability of high-priority tasks.

The application discloses a power cost prediction method and system based on a knowledge graph, and the method comprises the following steps: obtaining historical power data, wherein the historical power data comprises power generation side data, power transmission and distribution side data, power consumption side data, macroeconomic data and policy text data; constructing a power cost knowledge graph based on the historical power data; performing a message passing operation of a graph neural network layer by layer along the direction of the hierarchical structure to generate embedding features of each node; and inputting the embedding features of a target node into a prediction network to output a power cost prediction result. By constructing a cost transmission hierarchical structure, the application explicitly embeds the formation mechanism of the power cost in the industrial chain into the knowledge graph, realizes differentiated and accurate modeling of the cost along the transmission path, and can accurately quantify the contribution of different transmission links to the terminal cost.

This invention relates to the field of artificial intelligence technology, and discloses an AI-based interface interaction preloading method and system. The method includes: collecting operation commands executed by the user on the maintenance and diagnosis interface to generate a user operation path sequence; performing path matching analysis between the user operation path sequence and the diagnostic process knowledge graph to obtain a candidate path set; selecting the downstream node with the highest path weight value in the candidate path set as the target node, and performing scenario-based preloading on the target interface of the target node to obtain a preloaded scenario package; monitoring whether an interface switching command to switch to the target interface is issued, and when the command is detected, calling the preloaded interface instance to the maintenance and diagnosis interface, and simultaneously activating the operation context instance; and interactively binding the data items of the auxiliary dataset with the interactive elements on the target interface to construct the interactive target interface. This invention can improve the responsiveness and operational consistency of interface switching in the maintenance and diagnosis system.

This disclosure relates to the field of computer technology and discloses a method, apparatus, computer device, and storage medium for identifying associated communities. The method includes: acquiring each node in a social network and the atomic relationships between them; for any target node among the nodes, determining the graph embedding information corresponding to the target node based on the atomic relationships; constructing a social graph between the target node and all other nodes according to the graph embedding information; and identifying the target associated community from the social graph based on the node relationships in the social graph. By implementing the technical solution of this disclosure, multi-path information between nodes can be considered, facilitating better capture of complex relationships and features between nodes, and fully utilizing the graph data structure information of the social graph to achieve the identification of associated communities.

This application discloses a method, apparatus, device, and readable storage medium for thinking chain retrieval. The method includes: obtaining keywords from a first question input by a user; searching for target nodes matching the keywords at the business scenario layer of a first knowledge graph, and filtering multiple first candidate paths based on the target nodes and user profiles; calculating the semantic similarity between each first candidate path and the first question, obtaining multiple second candidate paths with similarity higher than a first threshold, and determining K high-scoring paths from the multiple second candidate paths based on the node access frequency and time decay factor of each second candidate path; and performing variable anchor point identification and logical coherence verification based on the K high-scoring paths to obtain the final thinking chain. This invention rapidly locates multiple first candidate paths through business scenarios and user profiles, and combines semantic similarity and path optimization to achieve precise targeted retrieval, thereby improving the accuracy of thinking chain retrieval results.

The present application relates to the technical field of computer, disclose a kind of node legend identification method, device, equipment, storage medium and program product, wherein, method includes: obtaining the candidate graph element to be identified in pipe network drawing drawing;From candidate graph element, obtain the target graph element set corresponding to target legend type;According to the identification rule corresponding to target legend type, identify target node legend in target graph element set;According to the geometric line of target legend type and target node legend, determine the feature information of target node legend.The present application realizes the automatic identification of node legend, improves the identification efficiency of node legend.