154 results about "Server allocation" patented technology

The invention discloses a load balancing method and a related device, and relates to the field of computers, and the method comprises the following steps: obtaining hash ring data; wherein the Hash ring data reflect the mapping relation between the server and the virtual nodes located on the Hash ring; adjusting the number of virtual nodes of the server based on the current load value of the server and a weight coefficient corresponding to a service scene to which the current service request processed by the server belongs; wherein the current service request is distributed to the server based on the Hash ring data; on the basis of the adjusted number of virtual nodes, Hash ring data are updated; and based on the updated Hash ring data, performing server allocation on the new service request, and sending the new service request to the allocated server for processing. According to the invention, the operation response is more timely while the load balancing is realized.

A method comprising receiving, via a server, a distress signal from a first drone determining among a plurality of drones performing mission tasks, is experiencing distress and has entered a mitigation procedure responsive to the mitigation procedure to include the content received from the first drone; and assigning, via the server, a new drone to replace the first drone as a follower drone or a leader drone.

According to the offshore edge computing task unloading and resource allocation method based on air-sea cooperation, the task queuing model is established, the optimization target model with the purpose of maximizing the average throughput of the system is established, and whether the unmanned aerial vehicle unloads the task to the server or not is determined after solving; when the unmanned aerial vehicle does not distribute the task to any server, the optimal computing resource distribution amount of the unmanned aerial vehicle end is obtained; when the unmanned aerial vehicle allocates the task to the server, the optimal transmission power of the unmanned aerial vehicle unloading task and the calculation resource proportion allocated to the unmanned aerial vehicle by the server are obtained, and then task unloading and resource allocation of the offshore edge system based on air-sea cooperation are completed. According to the invention, through accurate scheduling and efficient cooperation of resources of the unmanned aerial vehicle and the server side, inherent limitations of marine equipment on data acquisition breadth and calculation processing depth can be broken, maximization of the average throughput of the system is realized, and meanwhile, a firm and flexible technical support is provided for efficient processing of various marine tasks.

A system and method for assigning hosts to provide Cloud desktops to client devices of users is disclosed. The system includes available host servers for providing Cloud desktops to client devices of users. A monitoring service is coupled to the available host servers. The monitoring service collects usage data from the users of the Cloud desktops. A host allocation engine is coupled to the monitoring service and available host servers. The host allocation engine receives a request for a cloud desktop from a user. The host allocation engine determines usage data for the user requesting the cloud desktop. The host allocation engine allocates a host server from the available host servers based on the usage data of the user.

The embodiment of the invention discloses a locking method, terminal equipment and a computer program product, and belongs to the technical field of terminals. The method comprises the steps that n-tuple information is acquired, the n-tuple information at least comprises a device identifier and an authorization code, the authorization code is distributed by a locking server, both the device identifier and the authorization code have uniqueness, and n is larger than or equal to 2; sending a contract state query request containing n-tuple information; receiving a contract state query result sent by the locking server, wherein the contract state query result is obtained based on the n-tuple information; and under the condition that the contract state query result indicates that the n-tuple information does not pass verification, carrying out machine locking operation.

The e-commerce data processing method comprises the following steps: selecting an e-commerce product to form first e-commerce order information according to a hidden user identifier allocated to a first user by an e-commerce platform server; generating a first encryption factor; encrypting at least part of the first e-commerce order information according to the first encryption factor to obtain first ciphertext data; acquiring a first user public key corresponding to the first user hidden user identifier from an e-commerce data protection system server; wherein the first user private key is managed by a first user, and the first user public key and the first user private key generate a first user decentralized identity identifier; encrypting the first encryption factor according to the first user public key to obtain first encryption factor ciphertext data; the first ciphertext data and the first encryption factor ciphertext data are uploaded to an e-commerce data protection system server side and are stored corresponding to the first user public key; hash values of the first ciphertext data and the first encryption factor ciphertext data are linked to a block in the block chain, and are correspondingly recorded with the first user decentralized identity identifier.

The invention relates to a test system for a lithium ion battery, and the system comprises a terminal device which is used for sending a test request to a server, the test request comprises a test schedule, and the test schedule comprises a plurality of test tasks; the server is used for responding to a received test request sent by the terminal equipment, and distributing a corresponding test task to each control equipment based on a test schedule in the test request; the control equipment is used for responding to the received test task distributed by the server, controlling the connected test equipment to execute a corresponding test process according to test type information, test equipment information and test working condition information in the distributed test task, obtaining test result data of the distributed test task, and sending the test result data to the server; the test result data is used for calculating kinetic parameters of the battery sample. Therefore, high-throughput concurrent testing of testing tasks of a large number of battery samples can be realized in a streamlined and automatic manner, and the efficiency of acquiring kinetic parameters of different sample batteries can be improved.

The embodiment of the invention discloses a file cross-network transmission method and device and a computer readable medium. A specific embodiment of the method comprises the steps of determining file structure information of a to-be-transmitted file in response to the fact that the to-be-transmitted file passes compliance check; determining whether a target file template exists or not according to the file structure information; performing file content separation on the to-be-transmitted file to obtain a target file; generating a chained invisible watermark according to a template identifier corresponding to the target file template, meta-file information corresponding to the to-be-transmitted file and the target file; embedding the chain type invisible watermark into the file to be transmitted to obtain a watermark embedded file; according to the meta-file information, a ferry server matched with the to-be-transmitted file is distributed through a distribution server; and according to the ferry mode, the file embedded with the watermark is transmitted to an external storage space in a cross-network manner through the ferry server. According to the embodiment, efficient and safe cross-network data transmission is realized.

Described herein are systems, methods, and other techniques for assigning high-performance workloads to compute nodes in a compute infrastructure. The workloads are received at a workload scheduler. Each workload of the workloads includes a workload specification indicating requested resources to run the workload. Server statuses indicating resource usages and availabilities at the compute nodes are generated by a set of server agents running on a set of servers containing the compute nodes. The server statuses are sent from the set of server agents to the workload scheduler. Server assignments for the workloads are generated by the workload scheduler based on the requested resources and the resource usages and availabilities. Compute node assignments for the workloads are generated by one or more of the set of server agents based on the requested resources and the resource usages and availabilities.

The invention provides a channel resource allocation method of a power local wireless network based on WAPI, and belongs to the technical field of power networks. Comprising the following steps: an access point reports information of a target terminal device which applies for channel resources and is accessed or disconnected to a slave chain of an edge computing layer; the slave chain sends target terminal equipment information and a service priority query request to a main chain of the distributed core layer; the main chain obtains to-be-allocated terminal equipment according to the target terminal equipment information and the access point historical resource allocation scheme, and sends the to-be-allocated terminal equipment and the corresponding service priority to an allocation server of the slave chain; the distribution server carries out channel resource distribution calculation according to the service priority of the terminal equipment, obtains a resource distribution scheme, and sends the resource distribution scheme to the slave chain; the slave chain issues the resource allocation scheme to the corresponding access point; and the slave chain reports the resource allocation scheme to the main chain. According to the invention, the proper resource unit can be allocated according to the service type or the service priority of the terminal equipment.

The application provides a server distribution method, system and device, a server and a storage medium, and relates to the technical field of computers. The method comprises the following steps: receiving a game distribution request of a first game client sent by a role server; determining game servers with an online player number not reaching a preset player number threshold as multiple idle game servers from multiple game servers; determining a game server with the largest online player number as a target game server from the multiple idle game servers according to the online player numbers of the multiple idle game servers; sending an identifier of the target game server to the role server; the identifier of the target game server is used to instruct the role server to inform the first game client to establish a communication connection with the target game server according to the identifier, so that the target game server executes game logic corresponding to the first game client. The application can improve the game experience of players by improving the aggregation degree of players.

Apparatus and corresponding methods for distributing tasks across various processing devices, such as servers, of a distributed computing environment are disclosed. The embodiments can batch workloads and distribute the tasks to the various processing devices for processing such as to reduce an overall processing time and costs. The embodiments may, additionally or alternatively, monitor actual processing times, and reassign tasks that fail to complete within an expected time frame.

Technologies and techniques for providing specific connectivity quality of service along a route of a vehicle comprises, for at least one vehicle: transmitting route information and a definition of a minimum required connectivity quality of service to a service management server by user equipment of the vehicle; managing and reserving the required connectivity quality of service along the route by allocating communication resources using the service management server, which interfaces with a plurality of mobile network operators; and providing feedback to the user equipment regarding the availability of the required connectivity quality of service. The system includes a service management server and user equipment configured to execute the method. This system ensures specific connectivity quality of service, enabling advanced applications, such as remote driving, by dynamically managing resources based on vehicle routes and operator availability.

The present invention provides a multi-center redundant arbitration method and system based on a SCADA system, including: a server allocation sequence number step: the server will be allocated a sequence number according to the priority; a master-slave switching step: the active node is switched according to the priority of the server sequence number, and the server with a higher priority is selected as the current active node; at the same time, all inactive nodes will count the received heartbeats, and a voting request will be initiated to all nodes after a delay, and a new active node will be selected. The present invention uses a reliable third party to perform arbitration, so that when the redundant network card goes offline, the multi-network card server can determine that it is the local machine that goes offline instead of the active node, thereby maintaining the inactive node and avoiding the related services from affecting the applications of other servers through the remaining network cards due to the change of the redundant state.

The invention provides an unmanned ship navigation control method based on a virtual navigation control service. The unmanned ship navigation control method based on the virtual navigation control service comprises the following steps: acquiring unmanned ship navigation data according to a sensor preset on an unmanned ship, and sending the unmanned ship navigation data to a virtual navigation control server; different virtual navigation control modules are distributed through the virtual navigation control server according to the unmanned ship navigation data and received task requirements; and the virtual navigation control module obtains a navigation control instruction according to the navigation data of the unmanned ship and a preset route planning strategy, and sends the navigation control instruction to the unmanned ship through the virtual navigation control server so as to control the navigation of the unmanned ship. The route planning of the unmanned ship is calculated through the virtual navigation control module, navigation of the unmanned ship is controlled based on the route planning, and the technical problems that in the prior art, no calculation resource configuration is reasonable, and navigation control of the unmanned ship is poor in adaptability and slow in response are solved.

The invention discloses a resource adaptive federated learning-based multi-modal knee osteoarthritis auxiliary diagnosis method, which comprises the following steps that: a server initializes a knee osteoarthritis diagnosis-oriented multi-modal model, and pre-deploys a low-rank adaptive plug-in in a trainable layer; after the client reports local hardware resources, the server allocates a resource echelon and issues a two-dimensional cutting proportion; the client performs depth and width cutting on the model based on the cutting proportion, activates a LoRA module of a specified layer, loads local knee joint X-ray images and clinical text data to perform multi-modal training, and establishes semantic mapping of image features and clinical description; the client only uploads and activates the LoRA parameter update quantity, and the server optimizes the global model through weighted average aggregation and key layer compensation; and in combination with real-time resource monitoring and knowledge distillation, the learning ability of the model for key discriminant features of knee osteoarthritis is enhanced. According to the method, heterogeneous computing power equipment is covered through a resource adaptive mechanism, and the participation efficiency of basic medical institutions is improved.

The application designs a self-adaptive mobile edge computing unloading and resource allocation method, mainly including the following steps: a user sends a task unloading request; a mobile edge computing system calculates the best unloading strategy of the task according to the data size of the task, the wireless channel information where the user is located and the available computing resources of each edge server in the mobile edge computing system and notifies the user; the user connects a base station according to the notified unloading strategy and sends a task; after the base station receives the task, the base station sends the task to an edge server connected to the base station, the edge server allocates resources to process the task; the edge server sends the processed result to the base station, and the base station returns the processed result to the user. The application adopts a deep reinforcement learning method, obtains the best task unloading scheme at the current time according to historical system information, and reasonably allocates resources to process the task under the limitation of task delay, so that the consumption of resources is reduced and the utilization rate of the edge server is improved.

The invention discloses a clustering federated learning method based on a neural network model structure, and belongs to the technical field of federated learning. The method aims to solve the problem of model performance reduction caused by data isomerism, system isomerism and data drift in federated learning. According to the technical scheme, the method comprises the steps that S1, a server initializes the same initial model for each client; s2, the client receives the model distributed by the server, after pre-training, pruning is performed on the model by using a model pruning technology, and a pruned model and a model structure mask are generated; s3, the client continues to train the pruned model, and uploads the local model and the mask to the server; s4, the server receives the model and the mask uploaded by the client, detects data drift of the client, and performs clustering through a similarity algorithm or determines whether re-clustering is needed or not according to the consistency of a model structure; the method has the advantages that the accuracy of the model can be remarkably improved in a non-independent identically distributed data scene, and meanwhile, the clustering calculation overhead can be reduced.

The embodiment of the application provides a game server distribution method, system and storage medium, and relates to the technical field of computers, which can be executed by an agent server and comprises the following steps: receiving a distribution request of a GS sent by a target PaaS server, the distribution request of the GS being triggered when a login request of a target client is detected, the target client being any client in at least one client; then checking whether there is idle server information of the GS in a queue, the queue storing the idle server information of the GS; if there is the idle server information of the GS in the queue, obtaining the server information of the target GS from the queue and storing the server information in a preset storage space, so that the target client establishes a connection with the target GS based on the server information of the target GS. The game server distribution method, system and storage medium provided by the embodiment of the application can reduce the waiting time of the client, thereby improving the user experience.

The embodiment of the invention provides a communication method and device, and relates to the technical field of networks, and the method comprises the steps: first network equipment and other network equipment except the first network equipment in a local area network carry out main equipment election; if the first network device is elected as a main device, the first network device is configured with a DHCP server function, and a DHCP server backoff function is started, the first network device detects other DHCP servers; after other DHCP servers are detected, the dynamic IP addresses distributed by the other DHCP servers take effect, and the DHCP server function is closed; and if the other network devices are elected as slave devices, the other network devices are configured with the DHCP server function, and the DHCP server backoff function is started, the DHCP server function is closed, and the DHCP server backoff function is closed. According to the scheme, automatic networking during startup under the condition of multiple gateways can be realized, and zero-configuration opening of a user can be realized.

The invention provides a dyeing method for flow detection, a flow detection method, equipment and a medium, and relates to the technical field of communication. The dyeing method for stream detection comprises the following steps: generating a target data message according to acquired dyeing information and a to-be-detected data message; wherein the dyeing information comprises a data stream field, the data stream field is filled with an equipment identifier of the first detection node and a stream identifier allocated to the first detection node by the server, and the stream identifier is used for identifying a service stream corresponding to the to-be-detected data message processed by the first detection node; and sending the target data message to the second detection node. According to the embodiment of the invention, the method can guarantee the compatibility with an original detection system, does not need to increase extra system overhead, and improves the flow detection efficiency.

The invention discloses an environmental sanitation task intelligent management and control method and system based on a handheld interphone, and the method comprises the steps that each environmental sanitation worker carries an interphone for patrol, and the interphone is provided with a camera and a GPS positioner; the interphone finds the environmental sanitation task and reports the environmental sanitation task to the cloud server by adopting one or more methods of video analysis snapshot, preset condition automatic snapshot and manual active shooting; the cloud server performs overall scheduling according to the task information carried by each environmental sanitation task in combination with the environmental sanitation information of each environmental sanitation worker, and distributes each environmental sanitation task to the corresponding appropriate environmental sanitation worker; an interphone carried by an environmental sanitation worker receives the environmental sanitation task allocated by the cloud server; a sanitation worker presses a task start button on the interphone, goes to a designated position to complete the sanitation task according to prompt information of the interphone, and reports the task to the cloud server. According to the invention, discovery, reporting, intelligent guiding, intelligent planning and cooperative scheduling of environmental sanitation tasks can be realized.

The application provides a network proxy system and method divided by regions, the method divides a server into multiple VPS, acquires proxy server IP of different regions, and stores to a database, further provides an interface, when a calling party accesses a site across regions, through fixed interface parameters, acquires the region number of the calling party accessing the site across regions, extracts any one proxy IP corresponding to the region number of the calling party accessing the site across regions from the database as a forwarding server IP, and assigns the forwarding server to the calling party, so that the calling party forwards messages through the forwarding IP. The system and method support the calling party to report invalid proxy IP to trigger a dialing mechanism, can acquire a new proxy IP, well guarantees the operation stability of the calling party, and can bind a fixed proxy IP for the calling party, so that the calling party can reduce the database query when acquiring the forwarding server again, and reduces the occupation of system resources.

The invention relates to a semantic-driven collaborative edge computing double-time-domain caching and computing power collaboration method and system, and belongs to the technical field of mobile communication and edge computing crossing. Aiming at the problems of black box of semantic fidelity, insufficient space coupling among multi-edge servers, difficulty in dual-time scale resource collaboration and the like in the collaborative multi-antenna semantic perception MEC system, the method comprises the following steps of: establishing an uplink MEC system model, and defining semantic perception QoE indexes of coupling semantic similarity and end-to-end delay; according to the method, an optimization problem taking long-term reward maximization as a target is constructed, an SCALE framework is designed, long-term cache placement is optimized by combining a DDPG agent enhanced by a long short-term memory (LSTM) network, and short-term semantic compression, task unloading, server allocation and joint power computing resource optimization are processed by a real-time PPO integrated decision RAPID algorithm. Simulation results show that the method is superior to an existing baseline scheme in the aspects of convergence speed, long-term reward, average QoE and cache switching overhead, and the system performance and the resource utilization rate are improved.

The present disclosure relates to a method and device for implementing a headless browser cluster. The method comprises: assigning a corresponding accessible IP address to each server in a plurality of servers; wherein the plurality of servers form a cluster; installing a headless browser in each server; after receiving a headless browser starting instruction, starting a preset number of headless browser instances, the started headless browser instances forming a headless browser instance pool; after detecting that a headless browser instance acquisition instruction is received, acquiring a target headless browser instance available from the headless browser instance pool; acquiring a network socket endpoint address of the target headless browser instance; and sending the network socket endpoint address of the target headless browser instance to a client.

A System and method for producing K-Net devices and improving a K-Net communication network on which the devices run to provide electronic services. A divided Net Extender extends the wireless network. K-Net devices utilize a modem card as an internal or external modem over Wi-Fi, cellular, millimeter wave and microwave bands. A processor on a circuit board coupled to a G chip or traffic controller chip embedded with a network control protocol and coupled to a dedicated cache memory manages networking of mobile devices and autonomous vehicles from a Local Area Network (LAN) manager device. A geographical internet protocol (Geo IP) enhances security. The modem connects to a wireless router device that runs a card control protocol, wherein the router plugs into the LAN manager device, connecting to internal and external DNS servers, GIP allocation and authentication servers. Devices are self-powered by alternating current generated from a direct current.

The invention relates to the technical field of reading content recombination, and discloses a reading content recombination method based on reading consistency, which comprises the following steps: executing a server task allocation strategy, and allocating task load for each server based on feature data; executing the user reading behavior modeling strategy, extracting an embedded vector of the historical reading behavior sequence, and extracting user reading preferences according to the embedded vector; if the article currently read by the user belongs to the reading preference of the user, executing an article paragraph semantic analysis strategy, extracting semantics of each paragraph in the article, and calculating a coherence score of the article; if yes, executing a user preference calculation strategy, and calculating an interest map according to social behaviors and emotional states of the user; executing a dynamic recombination strategy, and dynamically calculating an optimal next recommended paragraph; and obtaining the click rate and the stay duration of the user, executing a feedback optimization strategy, and optimizing the selection of the recommended paragraph. The intelligent and personalized reading experience is provided, and the reading interest of the user is increased.

The application provides a dynamic load balancing method based on a long connection, relates to the technical field of micro services, and periodically receives a machine running index set uploaded by a service end cluster according to a running index receiving period through a coordinator; the coordinator receives a service end address request of a client, and activates a strategy decision maker in the coordinator according to the service end connection request; a load balancing strategy is obtained by calling in a load balancing strategy set; K groups of machine running indexes in the machine running index set are analyzed and compared to determine a link service end address, and a client and a service end based on a long connection network protocol are fed back to perform RPC calling. The technical problem that in the prior art, when a client server is distributed, the consideration dimension of a load balancing strategy is less, the flexibility of server distribution is weak, and the utilization rate of server resources is not high is solved. The technical effect that while the load balancing of a server cluster is borne, the utilization rate of server resources is better improved is realized.

The application discloses a resource self-adaptive federated learning multi-modal knee osteoarthritis auxiliary diagnosis method, which comprises the following steps: a server initializes a multi-modal model for knee osteoarthritis diagnosis, and pre-deploys a low-rank adaptive plug-in in a trainable layer; after a client reports local hardware resources, the server allocates a resource echelon and issues a two-dimensional clipping ratio; the client clips the model in depth and width based on the clipping ratio, activates a LoRA module of a specified layer, loads local knee osteoarthritis X-ray images and clinical text data for multi-modal training, and establishes semantic mapping of image features and clinical descriptions; the client only uploads an activated LoRA parameter update amount, the server aggregates and optimizes a global model through weighted average aggregation and key layer compensation; and the learning ability of the model to key discriminative features of knee osteoarthritis is strengthened in combination with real-time resource monitoring and knowledge distillation. The application covers heterogeneous computing power devices through a resource self-adaptive mechanism, and improves the participation efficiency of primary medical institutions.

The invention provides a load balancing method and device, computer equipment and a storage medium, and the method comprises the steps: responding to at least one task request of the same resource, determining the request number of the at least one task request, and according to the number relation between the request number and a set threshold value, determining a load balancing result of the at least one task request; determining a first identifier corresponding to the at least one task request; wherein the set threshold value corresponds to the processing capability of a server for processing the task request; determining a second identifier corresponding to the at least one task request, and generating a hash key value corresponding to the at least one task request according to the first identifier and the second identifier; wherein the second identifier corresponds to the identity information of the at least one task request; and performing server allocation of the at least one task request according to the hash key value.