50 results about "Server allocation" patented technology

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.

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 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 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 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 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 network attack defense method and device, equipment, a storage medium and a program product, and the method comprises the steps: determining at least one target server from a server cluster, wherein the target server is a server bearing a network attack; a plurality of target servers which can form an optimal server combination with the target servers are determined from a server cluster, and the optimal server combination is a server combination which is composed of the servers in the server cluster and has the highest computing power in all server combinations with the target servers; and distributing the plurality of target servers to the target servers to form an optimal server combination, and taking the optimal server combination as a defense server combination when the server cluster is attacked by the network. According to the embodiment of the invention, the optimal server combination can be accurately determined, and the optimal server combination is taken as a defense server combination when the server cluster is attacked by the network, so that the defense capability on the network attack is improved.

The embodiment of the application provides a kind of server updating method, service execution method and related equipment, it is related to communication, internet technology field;It can be applied to system upgrade maintenance, game service and the like scene.Server updating method includes: for the server of authentication version, version update operation is cyclically executed, until all the server of current version is updated to the server of target version;Version update operation includes: in the server cluster of current version, configure the server of authentication version;When receiving the server allocation request based on target service is initiated and determining that there are different versions of servers, the server of high version is determined as target server, and the target server is allocated to execute target service;If the target server is the server of authentication version and the response operation to target service meets the preset requirement, it is determined that the target server passes authentication and is updated to the server of target version.The server is updated without stopping service in the application.

The application provides a multi-MEC dynamic load balancing method based on an SDN controller, comprising the following steps: S1, constructing a network framework and building a coverage layer of the SDN controller; S2, when a vehicle is connected to a road test unit, transmitting an offloading task to a MEC server through the road test unit; S3, uploading characteristic information of the MEC server and the offloading task to the SDN controller; S4, performing offloading task distribution ratio calculation based on the uploaded characteristic information of the MEC server and the offloading task; S5, determining a decision result of the distribution ratio of each MEC server after the distribution ratio calculation is completed; S6, controlling each MEC server to perform offloading and calculation based on the decision result; and S7, dynamically adjusting the distribution ratio of each MEC server in the offloading and calculation process, so that the multi-MEC dynamic load balancing is achieved. The application solves the problem of unbalanced load among MEC servers.

The invention relates to the technical field of multi-task scheduling, in particular to a task collaborative management method and system based on templated workflows, and the method comprises the steps: constructing a target task collaborative template according to a plurality of test workflows, receiving a task collaborative management instruction, determining a target work task set based on the task collaborative management instruction, and constructing a target workflow and workflow data based on the target task collaboration template and the target work task set, performing multiple segmentation on the target workflow based on the workflow data to obtain a target work substream set, and performing server allocation on each target work substream in the target work substream set to obtain a task allocation server set. According to the method, the multi-task execution efficiency and stability can be improved, and resource consumption caused by unreasonable task scheduling is reduced.

This application provides a method and system for energy efficiency perception and load allocation of distributed network acceleration nodes, relating to the field of communication technology. The system inputs the server performance profile characteristics of each service server into an energy efficiency prediction model to determine the Nth estimated energy efficiency of each service server under different loads. The Nth estimated energy efficiency, Nth total service access volume, and Nth existing load of each service server under different loads are then input into a service access volume allocation model to allocate service sub-access volumes to each service server. Based on the Nth actual energy efficiency of multiple service servers, the Nth average actual energy efficiency of the multiple service servers is determined. If the Nth average energy efficiency is less than the (N-1)th average energy efficiency, and the difference between the Nth average energy efficiency and the (N-1)th average energy efficiency is greater than a set difference threshold, a reinforcement learning model is used to update the network parameters of the service access volume allocation model.

The embodiment of the invention provides a resource allocation method and device for terminal equipment, the equipment, a medium and a product. The method comprises the following steps: determining residual available resources in available resources of a server to obtain residual available resources of the server; determining a basic resource priority of the terminal device serving as the server device according to the residual available resources of the server; according to the network state of the terminal device, the variable quantity of the device connection quantity in a preset first time range, the variable quantity of data transmission in a preset second time range, the variable quantity of the number of times of terminal device connection in a preset third time range and a preset weight, the basic resource priority is adjusted, and the service end resource allocation priority is obtained; and reallocating server available resources and client available resources in the terminal equipment according to the server resource allocation priority. According to the resource allocation scheme of the terminal equipment provided by the embodiment of the invention, the resource utilization rate and the efficiency when the terminal equipment realizes own functions can be improved.

This application provides a page display method, apparatus, electronic device, and computer storage medium, belonging to the field of cloud security technology. In the embodiments of this application, in response to an operation triggered by a target object to access a target server, the client obtains a target verification identifier corresponding to the target server; wherein, the target verification identifier is obtained by executing an identifier acquisition script file issued by an attack protection device to access a third-party verification server; a first access request containing the target verification identifier is sent to the attack protection device, and after determining that the target verification identifier has passed verification, the attack protection device forwards the first access request to the target server; the response data of the first access request returned by the target server is received, and the corresponding display page is displayed according to the response data. The third-party verification server assigns a unique verification identifier, making it impossible for malicious software tools to forge the unique verification identifier configured by the third-party verification server, thereby further improving the security of accessing the server.

The application provides a load balancing method and system for an Internet of Things edge terminal. The load balancing method comprises the following steps: acquiring all devices for uploading data in the Internet of Things; classifying the data uploaded by the devices in units of devices to obtain data categories; summarizing the data generated by all devices in units of data categories, and acquiring the data volume corresponding to each data category and a data category set; distributing servers according to the data volume and the data categories to obtain the servers corresponding to each data category; and sending data to the corresponding servers according to the corresponding relationship between the data categories and the servers. The system comprises modules corresponding to the method steps.

The invention discloses an underwater virtual simulation interaction system, method and device, and relates to the technical field of underwater virtual simulation, the system comprises a simulation end, a media uplink channel, a media downlink channel and an asynchronous task distribution module, the simulation end comprises a plurality of robots Actors and a control server; the control server distributes a unique entity identifier Fish ID, and establishes a mapping relation between the Fish ID and a corresponding robot Actor in the control server; the frame stream component encodes the rendered picture and periodically sends the rendered picture in a preset frame format through a media uplink channel, and the frame stream component passively monitors and receives the superimposed image frame at a port paired with a media downlink channel; the asynchronous task distribution module delivers the asynchronous task to a main game thread of the UE5, and performs pose updating on the robot in the main game thread; the IMU simulation component is used for acquiring sensor data aligned with an image frame timestamp; according to the method, bidirectional closed-loop interactive underwater virtual simulation communication is realized, and the method is safe in thread, low in coupling, easy to expand and high in robustness.

The application discloses a disk symbol allocation method, a server and a system, and relates to the field of servers.In the process of allocating disk symbols by the server, the server acquires a preset disk symbol list or a self-defined disk symbol list, and allocates disk symbols corresponding to each hard disk in a plurality of hard disks in turn according to the order of the disk symbol list.The server allocates disk symbols at the software logic level by using the order of the disk symbol list, and does not need to manually change the cable wiring relationship in the server, so that the disk symbol adjustment mode is relatively more convenient.In addition, the disk symbol allocation is realized by using the order of the disk symbol list, so that different servers with the same cable wiring relationship can have different disk symbol allocation rules, thereby saving the difficulty of server case code maintenance.Further, the disk symbol list can be self-defined, so that the flexibility and applicability of disk symbol allocation can be improved.

The embodiment of the invention belongs to the technical field of cloud computing, and particularly relates to a data server distribution method, a cloud data management platform and a computing device cluster, and the method comprises the steps that a management server receives an IP query request and obtains data transmission time delay corresponding to each data server, the data transmission time delay is used for indicating the time delay when the data server sends data to the terminal. And the management server determines a first data server corresponding to the data transmission time delay in the plurality of data servers, and sends a first IP address corresponding to the first data server to the domain name server so as to instruct the domain name server to transfer the first IP address to the terminal, and the terminal requests data from the first data server according to the first IP address. By adopting the application, the operation cost of the management server can be reduced, and the efficiency of transmitting data from the data server to the terminal can be improved.

The invention provides an application access security control method and system and a computer readable storage medium, and the method comprises the steps: responding to an application access request from a client, and based on a preset server distribution strategy, distributing a target proxy server for the client in an available proxy server set; sending the application access request to an application resource server through the target proxy server and a zero-trust gateway connected with the target proxy server, and feeding back response data of the application resource server for the application access request to the client through the zero-trust gateway and the target proxy server; and when it is detected that the target proxy server does not have the client access flow, updating the IP address of the target proxy server according to a predetermined address dynamic updating strategy, and adding the target proxy server with the updated IP address into the available proxy server set again. According to the scheme, the proxy server capable of dynamically updating the address is dynamically allocated to the client, so that the network security is improved.

The invention relates to the technical field of smart home, and provides a method and a device for determining main and central network equipment, and electronic equipment. The method for determining the main and central network equipment comprises the following steps: receiving a priority number distributed by a cloud server; wherein the cloud server distributes a priority number for the central network equipment after receiving an access request of the central network equipment for the equipment interconnection local area network, and the equipment interconnection local area network adopts a main and standby central network architecture; according to the priority number, determining that the own priority is the highest in the device interconnection local area network by searching the priority numbers of other central network devices in the device interconnection local area network; and setting the device as a main central network device of the device interconnection local area network. According to the invention, automatic election of the main hub network equipment in the main and standby hub networks can be efficiently and reliably realized.

This invention provides a cloud instance reuse method based on physical machine local storage and object storage, relating to the field of virtualization technology. The method includes: adding a server to a platform and performing capacity planning upon addition; allocating N virtual instances of fixed specifications to the server; and allocating M additional free disks in the server's local storage to reserve storage space for unbinding operations of virtual instance disks. Based on the reserved storage space, when a virtual instance on the server is detected to meet inactivity conditions, the disk data of the inactive virtual instance is backed up to object storage, the association between the inactive virtual instance and the local disk is unbound, and the local disk is used as the unbound disk associated with the user. This invention achieves a balance between low-cost resource utilization and a high-quality user experience.

A computer implemented method correlates a local resolver to a client. The local resolver requesting an address to a resource from an authoritative domain name server is identified. An access pattern defining servers for accessing the resource over time slices is determined. The servers are assigned to the time slices and are configured to record requests to access the resource. Sending responses from the authoritative domain name server to the local resolver is initiated using the access pattern. Each response in the responses has the address to a server assigned to a current time slice during which a request for a new address is received from the local resolver. Whether the requests to access the resource from the client match the access pattern is determined. The local resolver is associated with the client in response to the requests matching the access pattern.

The present disclosure provides a method and system to render the push notifications in a communication device of a user. The present disclosure includes a software development kit. The SDK works even when the battery optimization in the communication are turned on. The present disclosure has dedicated server allocations. Servers use firebase cloud messaging APIs to send a notification to the communication device. The FCM servers send the notification downstream to the one or more communication device's operating system to broadcast an intent. This intent is listened by the firebase cloud messaging software development kit installed in a memory of the communication device. The SDK passes the notification payload to a notification service of an application and the notification is rendered. Further, the present disclosure prevents ANR errors and may enable the FCM service in devices that may have not supported FCM service previously.

The invention provides an energy-saving control method and system based on AI data center energy consumption analysis, and belongs to the field of energy-saving control. The problems of high energy consumption and low heat dissipation efficiency of the data center are solved; the method specifically comprises the following steps: acquiring historical energy consumption data of each server, and predicting an expected CPU utilization rate of each server by using an SARIMA model; constructing a power function of each server, and distributing actual power to each server according to the expected CPU utilization rate of each server; constructing a change function of the surface temperature of each server of the data center, and adjusting the working state of the heat dissipation system according to the heat dissipation mode of the data center and the actual power of each server; the historical energy consumption of the data center is obtained, processed and analyzed, the actual power of each server is adjusted to cool the working mode of the system, the energy consumption of the data center is reduced, and the operation stability of the servers is improved.

The application relates to the technical field of intelligent computing centers, in particular to an address allocation method and device and related equipment. The method comprises the following steps: determining tenant information of a newly created tenant, wherein the tenant information comprises target computing server information allocated to the tenant and target tenant network segment information allocated to the tenant; instructing a target access device accessed by the target computing server to create a logical virtual interface and setting an IP address of the logical virtual interface as a gateway address of the target tenant network segment; allocating a target IP address to the target computing server from the target tenant network segment; and sending an address allocation instruction carrying the target IP address to the target computing server through a communication link between an agent program of the target server and the target computing server, so that the target computing server configures a target network card based on the IP address carried by the address allocation instruction, wherein the target network card is a network card accessing the target access device.

The invention provides a server position distribution method, electronic equipment and a storage medium, and relates to the technical field of server position distribution, and the method comprises the steps: for any to-be-distributed server QW, determining a CPU utilization rate corresponding to each preset moment in an operation cycle of the QW; determining server power corresponding to each preset moment in the operation cycle of the QW; determining a CPU temperature determination model corresponding to the QW at each preset moment; determining a model according to the server power corresponding to each preset moment of the QW in the operation cycle and the CPU temperature, and determining the CPU temperature corresponding to each preset moment of the QW in the operation cycle; according to the server power and the CPU temperature of each to-be-allocated server at the same preset moment, allocating each to-be-allocated server to a corresponding cabinet; according to the invention, cabinet resource waste and energy redundancy are avoided, risks of local overheating and the like can be avoided, and the space and heat dissipation resources of the cabinet are utilized to the maximum extent.