39 results about "Web cluster" patented technology

The present disclosure relates to computer-implemented methods, software, and systems for distributing task execution at nodes of a web service cluster. A request is sent from an interface gateway to execute a task at a first service node at a web service cluster. The first service node is one of a plurality of service nodes at the web service cluster. The first service node is selected to execute the task based on evaluating a resource status matrix defining resource statuses of the plurality of service nodes in the web service cluster and resource requirements of the task. A status response from the first service node is obtained after executing the task. The status response comprises a current resource snapshot of the first service node. The resource status matrix is updated based on the obtained status response from the first service node.

The embodiment of the invention discloses a terminal data management method and device, electronic equipment and a storage medium. Through a multi-terminal cluster management triggering mechanism, monitoring data, installation position data and equipment operation state data are classified, only core data are transmitted and matched with terminal identifiers, the transmission efficiency is improved, and redundant data occupation is reduced; secondly, a standby switching design of a main communication module and a virtual operator network cluster communication module is adopted, so that the problem of data interruption caused by link instability in a complex environment is solved; and meanwhile, the core data are uniformly reported to the multi-terminal management server, so that multi-terminal uniform management and control and real-time state control are realized, the problems of low reliability and poor management efficiency of the existing scheme are solved, and the reliability and management efficiency of terminal data management are improved.

The utility model provides a computing system, network equipment, a distributed network cluster system and a PCB computing board card, and the computing system comprises a mainboard which comprises a first expansion bus interface; the DPU waste chip acceleration card is connected to the first expansion bus interface, the DPU waste chip acceleration card comprises an independent CPU subsystem and an acceleration engine, the CPU subsystem is normal, the acceleration engine has a fault, so that the DPU waste chip acceleration card is used as a CPU daughter card of a single board, and the DPU waste chip acceleration card is connected to the first expansion bus interface. The CPU subsystem and the acceleration engine have power supply and clock systems that are independent of each other. According to the technical scheme provided by the invention, the inherent CPU subsystem in the DPU waste piece can be reused as a computing node of other equipment, so that the waste is reduced.

Techniques are disclosed for a computing system comprising processing circuitry having access to a storage device, the processing circuitry configured to: generate, by a network controller executing in a software defined network (SDN), an advertisement in a first network cluster executing within a container orchestration platform of the SDN, wherein the advertisement conforms to a routing protocol and comprises information identifying a network service executing in the first network cluster, wherein the network service exposes a backend of a network application to the container orchestration platform of the SDN; and broadcast, by the network controller and to a second network cluster executing within the container orchestration platform of the SDN, the advertisement in accordance with the routing protocol.

The present invention discloses a multi-domain and multi-level cross-chain network collaboration method and device, which includes: obtaining cross-chain network collaboration information, and calculating the cross-chain weight value of each cross-chain network for other cross-chain networks based on the collaboration information; dividing the cross-chain network into regions and levels according to the collaboration information to form multi-level activity functions, element sets and multi-region activity functions and element sets; treating each region as a multi-agent, using multi-agents to link the activity functions and element sets of multiple regions in the cross-chain network to form a multi-agent cross-chain network cluster divided by domains; inputting the multi-agent cross-chain network cluster divided by domains into a trained multi-agent model to obtain the optimal domain weight value of the cross-chain network in each region; inputting the multi-level activity functions and element sets into a trained hierarchical reinforcement learning model to obtain the optimal layer weight value of the cross-chain network at each layer; and determining the final cross-chain network collaboration result.

The invention discloses a computing resource scheduling method and system based on a computing power network cluster, and relates to the technical field of computing power network clusters, and the method comprises the steps: determining computing tasks matched with a plurality of computing power nodes according to the calculation amount of the plurality of computing tasks, the node load states of the plurality of computing power nodes and a resource scheduling system; and the matching accuracy of the plurality of computing power nodes to the computing task is ensured. Therefore, a calculation progress distribution diagram of the computing power network cluster is determined according to the calculation progress of the plurality of computing power nodes on the calculation task and the calculation modes of the plurality of computing power nodes; according to the computing progress distribution diagram of the computing power network cluster, the computing resources matched with the computing power nodes and the task states of the computing tasks, the optimization system of the computing resources is determined, the computing power of the computing power nodes in the computing process is dynamically optimized, further optimization of the computing resources is achieved based on the optimization system of the computing resources, and the computing efficiency is improved. And the computing power of the computing power node in the computing process is improved.

The technology generally relates to systems and methods for operating a distributed processing network. Tenser Processing Units (TPUs) may be interconnected to one another as part of a TPU cluster or pod. A number of TPU clusters may be connected to one another via inter-cluster switches to form a distributed network of TPU clusters. The inter-cluster switches may be configured to efficiently direct data transmissions from a first TPU cluster to a second TPU cluster based on identification of the data transmission's final destination and performing next-hop lookup operations. The inter-cluster switches may also be configured to perform cut-through operations and may be configured to have an internal non-blocking connectivity.

The embodiments of the present disclosure provide a network cluster data acquisition method and apparatus, the network cluster data acquisition method comprising: acquiring question text, the question text being for at least one network cluster; processing the question text on the basis of a target data processing model to generate response text, wherein the response text comprises answer text and step analysis text, the answer text is generated by the target data processing model on the basis of the step analysis text, and the step analysis text is generated by the target data processing model according to the question text and network cluster data corresponding to each network cluster. By using a target data processing model for data acquisition and processing, the present solution addresses the operational complexity and low efficiency of traditional methods, thereby achieving flexible multi-dimensional, multi-indicator responsiveness, and significantly improving the speed and accuracy of data acquisition.

The invention relates to the technical field of development, can be applied to financial network scenes, and discloses a cloud native network configuration method and device, equipment and a medium, and the method comprises the steps: storing full node information in a network cluster to a distributed key value storage system through a command line interface; performing self-defined plug-in connection with the distributed key value storage system in response to the operation and maintenance tool, and performing deployment marking on a target node in the network cluster to obtain a deployment label of the target node; and associating and deploying the cloud native network and the routing daemon process on the target node in the network cluster according to the deployment label of the target node, so that the target node realizes bidirectional transmission of routing information through component linkage. According to the process, the cloud native network configuration process needing manual operation can be converted into an automatic process capable of being repeatedly executed, so that the configuration process of the cloud native network adapts to the dynamic characteristics of the cloud native environment, and the flexibility of the cloud native network in the deployment stage is improved.

The invention discloses a private network cluster access system and method based on gRPC bidirectional authentication. The flexibility, the expandability and the operation and maintenance efficiency of the system are remarkably improved while the safety is guaranteed. The system comprises a container management module used for realizing container management services including receiving and processing of user requests, business arrangement and request forwarding, and user interface management; the cluster end proxy module is used for realizing active connection and registration, long connection maintenance, security instruction execution, HTTP request proxy, local authority control, log auditing and self-healing capability; and the agent management module is used for realizing cluster connection management, multi-cluster access mode switching, access strategy management, cluster end agent centralized management, state monitoring and alarming, instruction issuing and routing, security flow agent, authority control and authentication and whole-process encryption guarantee.

The present application relates to a cluster shared traffic threshold-based rate limiting method and apparatus, which are used for achieving effective bandwidth management and allocation among a plurality of nodes in a cluster and among a plurality of network card queues in the nodes. The method comprises: S1, a cluster manager issuing a cluster rate limiting threshold of a specified data stream to each node in a network cluster, and adding the forwarding process of each node into a multicast to share traffic information between the nodes, the cluster rate limiting threshold being acquired in the forwarding process of the node; S2, a node controller collecting statistics about the traffic of the specified data stream flowing into a network card per second, performing marking, sending out, by means of multicasting, a unique identifier of the node and network card traffic received per second to be shared to other nodes in the cluster, and on the basis of a multicast packet received from the cluster, calculating a node rate limiting quota threshold of the present node; S3, on the basis of each service core for processing traffic of queues in the forwarding process of the node and the bandwidth proportion of the total traffic of the nodes, a bandwidth allocation scheduler allocating a core rate limiting quota threshold of each node; and S4, on the basis of real-time data of the processed traffic of each queue in the nodes and the proportion of the total bandwidth of the nodes, a token bucket manager performing periodical and dynamical adjustment by means of a token bucket algorithm to obtain the proportion of each queue in a cluster rate limiting bandwidth, so as to limit the rate of the traffic passing through the queue.

This application discloses a method, apparatus, electronic device, and storage medium for transmitting startup files, relating to the field of computer technology. The method includes: when the node to be deployed in a network cluster is in the basic system environment, obtaining the node status of other cluster nodes, and then determining the number of peer service nodes based on the node status of the other cluster nodes. When the number of peer service nodes meets the peer service conditions, it indicates that there are enough nodes in the network cluster to provide peer upload services. Each first file request is sent to the corresponding peer service node. The node to be deployed can obtain the first file block returned by the peer service node using the peer service download mode. In this way, the upload bandwidth of all peer service nodes in the network cluster is effectively utilized, thereby effectively improving the download speed of startup files and enhancing deployment efficiency.

The present invention relates to quality of service (QoS) of services. The present invention provides a QoS handling system for one or more services, the QoS handling system comprising a plurality of QoS handling nodes arranged in a hierarchical tree structure having at least two levels of hierarchy. The highest level of hierarchy comprises one or more root nodes, each root node being associated with a set of service instances of the services. The lowest level of hierarchy comprises a plurality of leaf nodes, each leaf node being associated with one service instance, each leaf node being derived from at least one root node, and the service instances of the set of each root node being associated with the leaf nodes derived from the root node. Each leaf node can apply a local QoS policy to the associated service instance of each leaf node, and each root node can apply a first common QoS policy to the set of service instances of each root node.

The present invention discloses a distributed messaging and monitoring method for multi-cloud computing environments, comprising the following steps: S1, searching for a target service node for a client to connect to a network cluster; S2, adding a DNS service to a k8s cluster management tool, and using the DNS service to elastically scale out the client's optimal service node from each target service node. By searching for a target service node that meets the client's service needs and then finding the optimal service node among each target service node, the present invention reduces the data interaction process required for the client to directly connect to a remote service node, thereby reducing network latency in message consumption.

Techniques are disclosed for a computing system comprising processing circuitry having access to a storage device, the processing circuitry configured to: generate, by a network controller executing in a software defined network (SDN), an advertisement in a first network cluster executing within a container orchestration platform of the SDN, wherein the advertisement conforms to a routing protocol and comprises information identifying a network service executing in the first network cluster, wherein the network service exposes a backend of a network application to the container orchestration platform of the SDN; and broadcast, by the network controller and to a second network cluster executing within the container orchestration platform of the SDN, the advertisement in accordance with the routing protocol.

The invention discloses a boot file transmission method and device, electronic equipment and a storage medium, and relates to the technical field of computers.The boot file transmission method comprises the steps that under the condition that a to-be-deployed node in a network cluster is in a basic system environment, the node states of other cluster nodes are obtained, and determining the number of the peer-to-peer service nodes based on the node states of the other cluster nodes, and when the number of the peer-to-peer service nodes meets the peer-to-peer service conditions, indicating that there are enough nodes in the network cluster to provide peer-to-peer uploading services. And sending each first file request to the corresponding peer-to-peer service node. In the embodiment of the invention, the to-be-deployed node can obtain the first file block fed back by the peer-to-peer service node by using the peer-to-peer service downloading mode, so that the uploading bandwidth of all the peer-to-peer service nodes in the network cluster is effectively utilized, the downloading speed of the startup file can be effectively improved, and the deployment efficiency is improved.

The present disclosure relates to computer-implemented methods, software, and systems for distributing task execution at nodes of a web service cluster. A request is sent from an interface gateway to execute a task at a first service node at a web service cluster. The first service node is one of a plurality of service nodes at the web service cluster. The first service node is selected to execute the task based on evaluating a resource status matrix defining resource statuses of the plurality of service nodes in the web service cluster and resource requirements of the task. A status response from the first service node is obtained after executing the task. The status response comprises a current resource snapshot of the first service node. The resource status matrix is updated based on the obtained status response from the first service node.

The invention discloses a flow control device and method based on a request preprocessing mechanism. The device comprises a routing analysis module used for receiving a target request and analyzing the target request to obtain a routing result; wherein the routing result is used for representing a forwarding path of the target request; and the network cluster module is connected with the route analysis module, and is used for matching from a preset route set according to the route result to obtain a target route, and forwarding the target request to a target address through the target route. According to the technical scheme, refined flow control and efficient scheduling of client requests in a complex service scene or network environment can be realized.

The embodiment of the invention provides a configuration method of a network element cluster management system and related equipment, the network element cluster management system comprises a plurality of network element equipment, and the method comprises the following steps: obtaining a user configuration strategy, and obtaining at least one current configuration strategy from a data configuration table of the network cluster management system; performing configuration conflict detection based on the user configuration strategy and the current configuration strategy to obtain a conflict detection result; when the conflict detection result shows that the user configuration strategy does not conflict with all the current configuration strategies, calling a database interface, and adding the user configuration strategy into a data configuration table to serve as a new current configuration strategy; the control strategy is generated based on the user configuration strategy in the data configuration table, the configuration parameters of the control layer and / or the monitoring layer in the network element cluster management system are adjusted based on the control strategy, and the multiple network element devices are controlled and / or monitored according to the adjusted control layer and / or monitoring layer, so that the management reliability of the physical network element cluster is improved.

The present invention discloses a flexible and efficient distributed machine learning method in a dynamic scenario. The feature of this method is that it adopts a dynamic and efficient distributed training framework, realizes dynamic training node orchestration, divides the cluster nodes participating in distributed training into training nodes and supervisor nodes. On this basis, it analyzes the distributed training, constructs a network communication model in a dynamically changing network cluster, and uses a node task adaptive scheduling and allocation algorithm to achieve flexible scheduling of computing nodes. Compared with the prior art, the present invention has the advantages of efficiently and flexibly scheduling computing nodes in a distributed training cluster, improving the flexibility of the cluster in actual application scenarios, enabling the cluster to adaptively cope with dynamically changing training nodes, and being able to effectively alleviate the communication bottleneck problem in the case of large-scale clusters or complex network conditions, and improving the communication efficiency of the cluster.

The embodiments of the present application propose a transmission optimization method and related equipment for a network cluster system. The network cluster system includes multiple network clusters, and the network cluster includes multiple network nodes. The method includes: first, obtaining the link status of the communication link between each two network nodes, and obtaining a link connection matrix, which is used to characterize the connection status of each communication link; then, obtaining a network cluster scheduling model, inputting all link statuses and link connection matrices into the network cluster scheduling model, and obtaining an output link optimized transmission rate sequence and optimized traffic distribution ratio for each network cluster; finally, adjusting the transmission rate of each communication link in the corresponding network cluster according to each link optimized transmission rate sequence, and adjusting the traffic distribution of the corresponding network cluster according to each optimized traffic distribution ratio. When information is transmitted between network nodes in the network cluster system, the transmission resource utilization rate can be effectively improved.

This application discloses a cloud-native network resource allocation method, apparatus, device, and storage medium, relating to the field of resource management technology, and applied to a network twin service for cloud-native network clusters. The method includes: constructing a topology graph representing the dependencies between microservices using instance information of each microservice in the cloud-native network cluster; evaluating the resource usage of each microservice based on preset business experience service metrics and real-time user experience data, combined with the topology graph; the preset business experience service metrics are performance indicators generated based on historical user experience data of each microservice, and the real-time user experience data are user experience data of each microservice in real time; and performing resource allocation operations on each microservice according to the corresponding evaluation results. This combines network twin technology with cloud-native network clusters, allocating resources to each microservice with a user experience-oriented approach, which can accurately optimize resource allocation to improve user experience and resource utilization efficiency.

In one embodiment, a method includes generating, by a pod deployment tool, a security context profile, associating, by the pod deployment tool, the security context profile with a deployment rule, and associating, by the pod deployment tool, a vulnerability level with the deployment rule. The method also includes identifying, by the pod deployment tool, pod policies associated with a pod located within a cluster of a network and analyzing, by the pod deployment tool, conditions of the deployment rule using the pod policies. The conditions may be associated with the security context profile and the vulnerability level. The method further includes determining, by the pod deployment tool, whether to allow deployment of the pod within the network in response to analyzing the conditions of the deployment rule.

The invention provides an RU information acquisition method and a container scheduling method and device based on RU information. The RU information acquisition method comprises the steps that first equipment acquires information of RU equipment managed by schedulable nodes in a network cluster; and the first device sends the information of the RU device to a second device, so as to determine a target scheduling node in the schedulable nodes. Therefore, the capability of the existing second equipment is enhanced, the collection of the RU information is supplemented, and the node selection is performed on the to-be-scheduled container based on the collected RU information of the schedulable node, so that the condition that the to-be-scheduled container, especially the to-be-scheduled container of a high-real-time task, is scheduled to the node which does not meet the RU requirement of the to-be-scheduled container is avoided, and the scheduling efficiency of the to-be-scheduled container is improved. Compared with the prior art, the method can be better applied to container scheduling in a high-real-time RU demand scene, not only can proper nodes be more accurately selected for container scheduling, but also the overall reliability of the network, the user experience and the resource utilization efficiency can be improved.

A method is disclosed for network agility traffic steering at a centralized management entity in a network system, the method comprising: receiving load management information from a plurality of cells within a network cluster; evaluating the load management information to determine resource allocation for the plurality of cells within the network cluster; generating traffic steering instructions based on the evaluation; transmitting the traffic steering instructions to the cells within the network cluster to dynamically adjust resource allocation and network performance; and enabling clustering of cells through containerized microservices coordinated using Kubernetes, allowing for flexible and scalable deployment of network functions.

The application discloses a kind of open virtual network cluster test method, device, equipment and medium, it is related to cloud computing technology field, the method includes: in the north of open virtual network cluster Database configuration pressure model logic network, and the port in pressure model logic network is bound with the computing node in open virtual network cluster, completes to open virtual network cluster and pressurizes configuration;In the process of pressurized configuration, control node and computing node in open virtual network cluster are monitored, and the health degree information of cluster and the computing node monitoring information corresponding to computing node are acquired;The application can realize the automatic balanced pressurization of large-scale business data, so as to be able to show the health degree, performance and stability of large-scale OVN cluster under the stress of large-scale business data by monitoring control node and computing node in OVN cluster, facilitate subsequent performance bottleneck positioning analysis and tuning, save development and test cost.

The invention relates to the technical field of automobile electronic network management, in particular to a partial network cluster fault isolation and recovery mechanism conforming to the AUTOSAR standard. Through PNC fault detection, isolation and self-healing recovery steps, PNC-level fault rapid detection, intelligent isolation and automatic recovery are realized. In the detection stage, the state is monitored through an NM-PDU containing a PNI / PNSR bit, and a fault is judged in a multi-condition mode; in the isolation stage, a fault PNC is marked by a coordination node, an isolation message is sent by adopting a time-sharing grading strategy, and a process is synchronously closed; in the recovery stage, instantaneous, continuous and hardware-level faults are adapted respectively through a three-level progressive mechanism. According to the method, the limitation of traditional whole network management is broken through, PNC-level accurate fault control is realized, fault diffusion is avoided, the recovery efficiency and robustness are improved, the AUTOSAR standard is followed, parameters can be flexibly configured, multiple types of vehicle network scenes are adapted, and stable operation of the system is guaranteed.

The application provides a kind of computing power network cluster dynamic construction method based on constraint condition, the method includes: the information of registered computing power node and extract the computing task characteristics in task demand;According to the computing task characteristics, the ability of computing power node is evaluated, wherein the ability evaluation is obtained according to the characteristic function of computing task;The computing power node as task initiation node is as root node, and the next computing power node is evaluated and selected based on the reachable routing path of routing table and the constraint condition, wherein the evaluation value of branch computing power node is obtained by evaluation function according to the ability of evaluation node through the ability evaluation Vi;The evaluation value search fitting of branch computing power node that can be routed is carried out by level by level, and the best computing power multistage collaborative computing power node cluster for the computing task is obtained.The application also provides a kind of computing power network cluster dynamic construction system based on constraint condition.

The application discloses a data synchronization method of a heterogeneous database platform, and comprises the following steps: step one, frame data is acquired from the heterogeneous database platform according to the same data structure, and the frame data needing to be updated is identified, and the identified frame data is data to be synchronized; step two, the network cluster inside a platform where the current service is located is listened to, real-time deployment identification of a data plane agent is acquired from the network cluster, and the data range to be synchronized is calculated according to the implementation deployment identification; step three, a plurality of logical modules perform logical operation on the identified data range to be synchronized; the data synchronization method of the heterogeneous database platform avoids multiple synchronization of modified data between the client and the server through steps of acquiring synchronization data, setting a synchronization range, generating a data list, confirming the synchronization range, generating a synchronization command and the like, and improves the data synchronization efficiency of the heterogeneous database platform.