259results about How to "Load balancing" patented technology

The invention belongs to the technical field of computing power scheduling, and particularly relates to a computing power scheduling method in a distributed computing environment. According to the task scheduling method and system, the task needing to be processed at present is obtained through the task management system, the task is divided into the real-time task set, the priority task set and the common task set, and the related tasks are scheduled by adopting the corresponding scheduling strategy based on the set corresponding to the current task and the priority in the set. Resource scheduling and allocation in a distributed computing environment are realized; according to the method, through the steps of task classification, real-time resource monitoring, resource demand estimation, priority ranking, reasonable scheduling strategy, load calculation, resource migration and the like, the technical problem of computing resource scheduling and allocation in a distributed environment is effectively solved, timely processing of high priority is ensured, the system load is balanced, and the system performance is improved. And efficient utilization of resources and stable operation of the system are realized, so that the performance and reliability of the whole distributed computing environment are improved.

The invention discloses a power distribution system and method for a modular power supply, and aims to optimize power distribution through an intelligent means, improve the operation efficiency and stability of the system, and control a unit to check the states of all power supply modules, and evaluate the required total power output according to the load demand. Then, based on the capability of each module and a redundancy planning principle, an appropriate module combination is selected to meet the load requirement, and necessary voltage or current adjustment is carried out to ensure stable power supply. In the operation process, the control system continuously monitors the state and the overall performance of each module, and once a fault or an abnormal condition is detected, measures such as automatic switching to a standby module are taken immediately, so that the continuity and the safety of the system are ensured. In addition, through analysis and learning of historical data, the system can continuously optimize a power distribution strategy, and efficiency and reliability are further improved. And the ability of the system to deal with emergency situations is also enhanced.

The invention relates to the technical field of micro-grid distributed control, in particular to a micro-grid cluster distributed control method and system based on information interconnection. The method comprises the following steps: identifying a micro-grid cluster in a region, and extracting a plurality of micro-grid nodes; obtaining a plurality of microgrid operation state parameters; performing filtering standardization processing on the plurality of micro-grid operation state parameters, and constructing a micro-grid time sequence load change curve; discrete wavelet transform is carried out on the micro-grid time sequence load change curve, load fluctuation frequency trend mining is carried out, and micro-grid load fluctuation frequency trend characteristics are generated; carrying out data stream encapsulation on the load fluctuation frequency trend characteristics of the micro-grid, and constructing a cloud environment distributed computing framework; carrying out dynamic load demand calculation on the plurality of micro-grid nodes based on a cloud environment distributed calculation framework so as to obtain multi-period load demand prediction data of each micro-grid; according to the invention, high-efficiency microgrid cluster control with high energy utilization rate is realized.

The invention provides an HIVE grouping operation performance optimization and scheduling method based on key column partitioning, which relates to the technical field of data processing, and comprises the following steps: preprocessing data by constructing a key column distribution histogram and determining a key column partitioning strategy; optimizing data partitions based on resource evaluation and a dynamic programming algorithm; and calculating an optimal data transmission path by adopting a minimum spanning tree algorithm, and dynamically adjusting an allocation scheme. According to the method, HIVE grouping operation performance can be improved, dynamic balance of calculation loads is achieved, the problem of data skew is effectively solved, and large-scale data processing efficiency is improved.

The invention provides a task queue-based heterogeneous resource pool real-time scheduling method, which comprises the following steps of: splitting tasks transmitted at random moments into sub-task sets with topological relations, adding the tasks into a queue to open up a track, and sequencing and numbering according to priorities; the bottom layer chip is interconnected, a resource pool monitor is added to monitor the state of resources, a heterogeneous resource pool is constructed, and data is provided for a scheduling system; the scheduling system obtains the real-time states of the queue and the resource pool, monitors the occurrence of events, makes a dynamic decision, sets an optimization target, and schedules the required computing resources for the subtasks according to the priority in the queue. According to the method, the parallel computing advantage of the FPGA chip is fully exerted, high-speed communication in the heterogeneous resource pool is ensured, the resource utilization efficiency of the resource pool is greatly improved, the total task set execution time delay can be effectively reduced, and meanwhile, the heterogeneous resource pool can achieve the load balancing effect.

The invention discloses a hybrid expert model training system based on layer freezing, which relates to the field of model training and comprises an external memory, a freezing controller, a resource allocation controller and a plurality of computing devices. The external memory stores a data set for training the hybrid expert model; the freezing controller performs freezing decision making according to a path selected by an expert and sends the decision to the computing device for execution; the resource allocation controller executes a lexical grouping algorithm based on a frozen layer expert selection path to calculate expert popularity, and allocates computing resources according to the expert popularity; and the computing device stores the hybrid expert model, executes model training operation, executes a layer freezing task according to a layer freezing decision, and feeds back a freezing layer expert selection path and an expert selection path to the resource allocation controller and the freezing controller respectively. According to the method, training of the hybrid expert model is performed through a layer freezing technology, so that the training cost is reduced while the model precision is kept; computing resource allocation is carried out based on a word source grouping algorithm, and equipment load balancing is achieved.

The invention belongs to the technical field of data storage management, and discloses a distributed storage management system and method for cross-border e-commerce trade data. Comprising the steps of collecting cross-border e-commerce trade data; formulating a data classification rule based on the cross-border e-commerce trade data, and dividing the cross-border e-commerce trade data into a plurality of data subsets according to the data classification rule; constructing a heterogeneous coding graph for each data subset, and coloring the heterogeneous coding graphs to generate a corresponding heterogeneous coding data subset; distributing and storing the heterogeneous coded data subsets in a plurality of heterogeneous storage nodes according to a preset distribution rule, and grouping the heterogeneous storage nodes to obtain n heterogeneous storage node groups; a storage space is allocated for each heterogeneous storage node group, and distributed storage is completed; the method not only optimizes the data management process, but also provides a more reliable data basis for deep data mining and business intelligence analysis.

The invention provides a model training method and device, electronic equipment and a storage medium, and relates to the technical field of distributed computing. The method comprises the steps of dividing a context into context tasks based on structural analysis of a to-be-trained model; performing dependency relationship analysis on the context task to determine a task dependency graph; configuring priorities for the context tasks according to the task dependency graph, distributing the context tasks with the configured priorities to corresponding computing nodes, training according to an execution sequence, and obtaining load information and task execution information of the computing nodes; and according to the load information and the task execution information of the computing nodes, scheduling context tasks among the computing nodes, adjusting the execution sequence of the context tasks in the computing nodes, and outputting a training result. According to the real-time load condition of each node, task allocation is adjusted, load balancing of the computational nodes is realized, the utilization efficiency of computational resources is improved, a context task scheduling strategy is optimized, the utilization rate of the computational resources is improved, and the training time is shortened.

The invention provides a novel wireless blind detection method, device and equipment, a medium and a computer program product, and the method comprises the steps: converting a received air interface analog signal into time domain stream data, and sequentially carrying out the synchronous processing and fast Fourier transform of the time domain stream data, and obtaining frequency domain stream data; determining a candidate set space to be subjected to blind detection based on the frequency domain stream data; performing network identifier detection on the candidate set space to be subjected to blind detection to obtain a balanced data block corresponding to a target identifier; and determining a blind detection result of the air interface analog signal based on the balanced data block. According to the method, data stream processing and data block balancing are sequentially carried out on the air interface analog signal, and under cooperation of the data stream processing and the data block balancing, load balancing of blind detection of the new wireless passive positioning system is achieved.

The invention belongs to the technical field of power data transmission, provides a power data release path planning method and system based on load balancing, and provides a multi-path transmission strategy optimization method based on LSTM prediction and MPC, and the method comprises the steps: firstly constructing a multi-path data transmission state prediction model through LSTM; acquiring a data transmission state prediction value of each path at the next moment; then, inputting the predicted value at the next moment into the LSTM model so as to obtain the data transmission predicted quantity of a period of time in the future; and finally, on the basis of taking the set minimum release time as a target function, adopting rolling optimization and feedback correction of MPC to obtain each path transmission control quantity meeting the target function. The efficiency that the electricity consumption information acquisition system publishes the total acquisition data to the marketing business application system is improved, and a load balancing strategy is realized.

The invention is suitable for the technical field of power distribution network reconstruction, and discloses a snowflake power distribution network dynamic reconstruction method, system, device and medium, and the method comprises the steps: obtaining the basic data of a snowflake power distribution network in a research area; based on the basic data of the snowflake power distribution network in the research area, taking the minimum load balance degree and active power loss as targets, and constructing a load balance-oriented initial snowflake power distribution network dynamic reconstruction model; adopting a second-order cone relaxation technology and a large M method to convert the initial snowflake power distribution network dynamic reconstruction model into a target snowflake power distribution network dynamic reconstruction model of mixed integer second-order cone programming; solving the dynamic reconstruction model of the target snowflake power distribution network; and obtaining a snowflake power distribution network dynamic reconstruction strategy for load balancing based on a solving result. According to the method, the safety and economy of system operation can be considered, load transfer is carried out by comprehensively considering the load conditions of a plurality of superior substations and a plurality of feeder lines, so that the load balancing of the whole network is realized, the solving difficulty is low, and the decision-making time is short.

The invention belongs to the technical field of environmental protection, and discloses a method for rapidly and efficiently removing and recycling polycyclic aromatic hydrocarbon in water and a carbon-based adsorbent, the method comprises the following steps: S1, preparing a modified carbon-based adsorbent; s2, treating the polycyclic aromatic hydrocarbon pollutant wastewater; and S3, recycling the polycyclic aromatic hydrocarbon. According to the preparation method, H3PO4 or ZnCl2 is taken as an activating agent, and an activated carbon base material and an activating process are controlled, so that nano-micron multi-scale modified carbon-based adsorbent particles which are mainly prepared from a large number of long-strip-shaped gap mesopores which are communicated with one another and are approximately distributed in parallel and are auxiliary from a small number of discrete irregular groove-shaped mesopores are obtained; the long and narrow long-strip-shaped gap mesopores of the particle are matched with the irregular groove-shaped mesoporous structure, so that large-size polycyclic aromatic hydrocarbon can be quickly adsorbed and desorbed, and the balance of adsorption and desorption efficiency can be realized; on one hand, the adsorption and desorption efficiency of pollutants is improved, and on the other hand, the dosage of the adsorbent is greatly reduced.

A resource management system based on cloud computing relates to the technical field of resource management and comprises a cloud server, and the cloud server is in communication connection with a data acquisition module, a data storage module, a flow prediction module, a resource extension module, a resource activation module, a real-time flow monitoring module and a data aggregation module. The data acquisition module is used for acquiring data information; the data storage module is used for storing data information; the traffic prediction module is used for acquiring predicted data traffic of each edge equipment terminal; the resource extension module is used for presetting to-be-activated extension elastic computing nodes; the resource activation module calculates a pre-estimated activation time period of the node according to the determined expansion elasticity; the real-time flow monitoring module is used for judging whether a to-be-aggregated terminal is generated or not and determining to-be-aggregated flow of the to-be-aggregated terminal; the data aggregation module is used for generating an aggregated data packet and a verification list; and the cloud server is used for carrying out integrity verification on the aggregated data packet to ensure that the cloud server provides stable, safe and efficient cloud services.

The invention relates to a large-scale low-orbit satellite constellation network access node selection method, and belongs to the technical field of satellite communication. The method comprises the following steps: firstly, calculating positions and beam coverage conditions of high and low orbit satellites in an earth-fixed coordinate system at each moment through ephemeris, platform and load information of the satellites; secondly, designing an adaptability function according to the position relation between satellites, beam coverage, communication link channel quality and service demand conditions; and finally, a satellite suitable for serving as a network access node is selected from a large-scale low-orbit satellite constellation through a genetic algorithm. The method can be used for realizing efficient dynamic connection node selection in high-orbit and low-orbit satellite combined networking, so that functions of resource coordination, load balancing and the like among satellites are realized, and the deployment efficiency of a large-scale low-orbit satellite constellation is improved.

The invention relates to a session holding method and system based on Nginx and Lua. The method comprises the following steps: acquiring request information initiated by a client; analyzing the request information, and matching a corresponding request path based on location block information in an Nginx configuration file; if the request path is successfully matched, calling a Lua script in the location block and determining an adopted session holding strategy; and forwarding the request information to a specified back-end server according to the determined session maintenance strategy based on upstream block information in an Nginx configuration file, so that the same user request is routed to the same back-end server, thereby realizing session maintenance and load balancing.

The invention relates to the technical field of Internet of Vehicles data transmission, in particular to an intelligent networked automobile data transmission method and system.The intelligent networked automobile data transmission method comprises the steps that firstly, the interaction frequency degree of a user and an edge server is analyzed, and a demand user needing to be provided with a mobile server is screened; then analyzing the similarity between any two demand users, and grouping the demand users; setting a mobile server matched with the user group and planning a moving path of the mobile server according to the quantity distribution condition of the demand users in the user group and in combination with the matching index; the distribution of the demand users in each user group is adjusted by analyzing the moving trend difference of all the demand users in each user group in real time; and replacing the mobile server according to the residual electric quantity and the residual memory. According to the method and the system, the mobile server is used for assisting the edge server to process the vehicle data in each user group, so that the load of the server in the network is balanced, and efficient and stable data transmission is ensured.

The invention discloses a knowledge graph link prediction method and device, equipment and a medium, and relates to the technical field of knowledge graph link prediction, and the method comprises the steps: carrying out the entity type normalization processing and relation type normalization processing of an initial knowledge graph, and obtaining a normalized knowledge graph; grouping a plurality of target entity relationships corresponding to various target entities between every two entity types in the normalized knowledge graph to obtain a target number group of entity relationship samples; all target entity relationships in one group of entity relationship samples correspond to two entity types; the entity relationship comprises a head entity, a relationship type and a tail entity; uniformly distributing the target number of groups of entity relationship samples to computational nodes through a graph neural network to obtain a number of sample groups of computational nodes, and training the sample groups in parallel through a replica network to update graph neural network parameters, obtain a link prediction model and realize knowledge graph link prediction; a copy network of a graph neural network exists in each computing node.

The invention discloses a data center energy consumption optimization method based on a joint multi-objective optimization model, and the method is based on an SAC-Discovery algorithm, and comprises the steps: S1, selecting a server field, calculating the cooling power consumption of different server fields through employing a cooling power consumption model, or employing the cooling power consumption model and a dynamic pricing model, and carrying out the calculation of the cooling power consumption of different server fields; selecting the server field with the minimum cooling power consumption as an optimal server field; s2, selecting a server, calculating server newly-increased power consumption of different servers in the optimal server field by using a server power consumption model, or by using the server power consumption model and a dynamic pricing model, selecting the server with the minimum server newly-increased power consumption as an optimal server, and allocating a task to the optimal server; and S3, temperature regulation and control: calculating the temperature change of the equipment where the data center is located under the natural condition by utilizing a temperature control model and combining the cooling power consumption of the optimal server field and the server power consumption of the optimal server, and carrying out temperature regulation and control.

The invention discloses a double-tension-interval-based film feeding system for a multi-column packaging machine, which comprises a paper roll mechanism arranged at the bottom of a rack, and a paper film changing mechanism, a film feeding device, a swing shaft roller tensioning mechanism, a laser coding mechanism, a laser coding position adjusting mechanism and a damping mechanism are sequentially arranged at the downstream of the paper roll mechanism according to the film feeding direction, an encoder alarm device and a mark tracking photoelectric sensor mark position adjusting mechanism are further arranged on the machine frame. During working, the film feeding device pulls out a paper film from a paper film roll of the paper roll mechanism, and the brake shaft is braked by the brake block, so that the paper film forms a first tension interval between the paper roll mechanism and the film feeding device; the damping mechanism is located at the output end of the film feeding system, the paper film is wound in a swing tensioning roller shaft system of the swing shaft roller tensioning mechanism, and the swing tensioning roller shaft system droops under the action of gravity, so that the paper film forms a second tension interval between the film feeding device and the damping mechanism. The film feeding device can be matched with a multi-column packaging machine, can continuously feed films and is stable in tension.

The invention relates to an Internet of Vehicles computing power scheduling method based on a graph neural network and deep reinforcement learning, and the method comprises the steps: S1, collecting the state information of network entities in the Internet of Vehicles and the association information between the entities in real time, and constructing a dynamic time-space attribute graph; s2, inputting the dynamic space-time attribute graph into a pre-trained graph neural network encoder, and outputting a node embedding feature set containing a high-order topological relation and global graph embedding features through multi-layer message passing and feature aggregation; s3, the node embedding feature set and / or the global graph embedding feature are / is used as the state of a deep reinforcement learning agent and input to a strategy network, the computing power scheduling action at the current moment is output, and the computing power scheduling action comprises the steps of assigning an unloading target node for a to-be-processed computing task and distributing corresponding computing and communication resources; s4, dispatching actions are distributed to the corresponding network entities to be executed, environment feedback is collected and used for model updating and next round of dispatching, and the method has the advantages of improving dispatching efficiency, adaptability and system energy efficiency and the like.

The invention discloses a method for realizing resource service based on OAuth2 and a gateway, and relates to the technical field of information systems. Comprising the steps of 1, creating a resource manager component on a resource server based on an OAuth 2.0 Resource Server component and creating an API gateway component on the basis of SpringCloud Gateway, and 2, receiving an access token carried by a client through the API gateway component, verifying the validity of the access token, checking the permission range of the access token passing verification, determining whether the client has the right to access corresponding resources or not, and if the client has the right to access corresponding resources, executing the step 3, and if the client has the right to access corresponding resources, executing the step 4. Step 3, receiving an access token carried by the client through the resource manager component, verifying the validity of the access token, checking the permission range of the access token passing the verification, determining whether the client has the right to access the corresponding resource or not, if yes, forwarding the request to the correct resource server, and enabling the request to access the resource server; and if the check is passed, providing the requested corresponding resources through the resource manager component.

The invention discloses an intelligent dynamic load balancing method and system based on Clico-VPP. The method comprises the following steps: acquiring a load state index of a back-end service pod; according to the load state index and a service capability coefficient corresponding to the back-end service pod, calculating a dynamic traffic allocation weight of the back-end service pod through a weighted calculation model; based on a dynamic threshold synchronization strategy, synchronizing the state of a first hash ring located in a control plane to a second hash ring located in a forwarding plane; and based on the second hash ring, determining a target pod for the input data stream through consistency lookup. A double-loop decoupling architecture is provided, hash rings of a control plane and a forwarding plane are separated, a synchronization strategy is determined through a dynamic threshold value, decision intelligence and forwarding stability are both considered, and jitter is avoided; meanwhile, a dynamic weight-driven consistent Hash algorithm is adopted, a real-time load index and Pod service capability are fused to calculate a dynamic weight, and the dynamic weight is mapped into the number of virtual nodes, so that intelligent scheduling based on capability and maximum maintenance of connection during capacity expansion and contraction are realized.

The invention discloses a TTE integrated Ethernet test data time delay deterministic transmission method, which comprises the following steps that: a time synchronization core module is designed in a TTE protocol by utilizing an FPGA (Field Programmable Gate Array), and the module can process a received protocol control frame (PCF) to realize multi-node high-precision time synchronization; the state and other core signals are controlled by adopting a command-condition interaction method; a maximum matching algorithm combined with cut-off priority is used for generating a scheduling table, and the method comprises the eight steps of scheduling parameter input, period calculation, shortest path calculation, cut-off priority preprocessing, maximum matching sorting, message distribution, conflict detection and scheduling table loading. The FPGA is used for loading the scheduling table to the scheduling configuration module, Ethernet transmission is fused, periodic time slot distribution is carried out on links, and the stability of time synchronization and the time delay certainty of test data transmission are ensured. According to the method, the link load can be balanced on the premise of reducing the real-time data packet loss rate.

The invention provides an http request task queue concurrent processing method based on a UE engine, and relates to the technical field of network communication methods, and the method comprises the steps of request receiving, data acquisition, model construction, load prediction, request enqueue, task scheduling, task distribution, request response and the like. According to the method, the system load change can be accurately predicted, so that the processing of the request can be dynamically adjusted according to the actual state of the system load, and the instability caused by static scheduling in a traditional method is reduced.

The invention discloses a load balancing method and device, electronic equipment and a computer readable storage medium, the load balancing method and device are applied to a storage end which is in communication connection with each client through a physical network card, and the method comprises the following steps: determining whether to trigger a load balancing operation according to the total data traffic of each physical network card; if the load balancing operation is triggered, for any target client, determining all communication protocol connections between the target client and the storage end; according to all the communication protocol connections, determining a communication protocol connection set of each physical network card relative to the target client, and further determining data traffic of each physical network card relative to the target client; and determining a target physical network card according to the data traffic of each physical network card relative to the target client, and adjusting a communication protocol connection set of the target physical network card relative to the target client to realize load balancing. According to the technical scheme, load balancing among the physical network cards in the storage system can be achieved on the premise that normal operation of system services is not affected.

The invention relates to the technical field of power acquisition terminal data storage, and discloses a data hybrid storage method and device and computer equipment, which are applied to a memory database, and the method comprises the following steps: when electric energy meter reading data is written into a main database of a power acquisition terminal, receiving a data change notification sent by the main database; recording the meter reading data into a data cache by adopting a key value mode based on the data change notification; and preferentially querying the corresponding meter reading data in the data cache based on a preset data retrieval requirement, and if the corresponding meter reading data is not queried, switching to the main database to continue data retrieval. According to the method, a large amount of data is prevented from simultaneously accessing the main database at the same moment, so that the load of the main database is reduced, load balancing is realized, the memory database is preferentially queried during data retrieval, and the main database is switched to retrieve when the memory database cannot be queried, so that the access frequency of the main database is greatly reduced, and the data retrieval efficiency is improved. And the performance bottleneck during batch data query of the relational database is solved.

The invention discloses a mobile robot performance system popularity racking method based on an improved TabNet algorithm. Comprising the following steps: collecting historical order data; the historical order data is preprocessed; a TabNet algorithm is improved, and a popularity model based on the improved TabNet algorithm is constructed; training a popularity model based on an improved TabNet algorithm by using the preprocessed data set, calculating a popularity score of each SKU, and performing real-time data updating and dynamic recalculation of the popularity score; and according to the SKU popularity score obtained through calculation, goods are put on different goods shelf areas. The invention provides a novel mobile robot performance system popularity shelving method based on an improved TabNet algorithm, so that high-popularity SKUs can be effectively identified and shelved in a concentrated manner through dynamic feature selection and a popularity scoring mechanism, thereby improving the order line hit rate of an RMFS system and optimizing the sorting efficiency.

The invention provides a load balancing method and device, electronic equipment and a storage medium, relates to the technical field of network transmission, and can solve the problem of non-uniform distribution of network resources during data transmission. The method comprises the steps that the operation state of nodes and the operation state of links in a target network are acquired, the nodes comprise a local node and a remote node, and the links are links between the nodes; based on the address information of the target data, the operation state of the node and the operation state of the link, a target transmission path is determined, and the target transmission path at least comprises a local node and at least one far-end node; and sending the target transmission path to each node in the target transmission path so as to transmit the target data according to the target transmission path. According to the invention, load balancing of network resources can be realized during data transmission.

The invention relates to a fine-grained multi-operator parallel scheduling method and system based on a heterogeneous data stream architecture. The method comprises the steps that an operator topological graph is established according to operators obtained through decomposition and related to the task scheduling process, and an operator data flow graph is constructed; analyzing a data dependency relationship among operators in a task according to the operator topological graph, grouping the operators based on the data dependency relationship and the runtime characteristics of the operators, and maintaining a corresponding scheduling queue for each operator group; scheduling the parallel operator group according to the scheduling queue, and allocating computing resources for the parallel operators based on runtime characteristics; performing data flow diagram dynamic reconstruction on the original operator data flow diagram of the parallel operator group according to the operator data flow diagram; and after the PE data flow array finishes executing the operators in parallel, checking whether each scheduling queue is empty or not, if so, ending operator scheduling, and otherwise, continuing to schedule the parallel operator group according to the scheduling queues. According to the method, the access to the SPM can be reduced, and multi-operator efficient parallel scheduling is realized.

The invention provides a method and a device for deploying a perception service, and a storage medium. The method comprises the following steps: a first SF receives a first perception request sent by a control plane network element; and determining one or more second SF based on the first perception request, and sending a perception data processing rule to the one or more second SF.