380 results about "Cluster systems" patented technology

The invention relates to the technical field of artificial intelligence, in particular to an intelligent operation and maintenance system of a cluster system based on reinforcement learning. The system comprises a cluster modeling module, a multi-dimensional situation awareness module, an intelligent decision planning module and an efficiency evaluation module, and realizes autonomous operation and maintenance and dynamic management of the cluster system in a complex environment in combination with various reinforcement learning algorithms. According to the system, reinforcement learning algorithms such as a deep Q network (DQN), Actor-Critic Monte Carlo Tree Search (AC-MCTS) and distributed near-end strategy optimization (DPPO) are introduced to optimize a maintenance time sequence and a strategy, so that the fault rate and the operation and maintenance cost of the system are reduced; the system function is quickly recovered, and the performance loss caused by sudden failure of the system is reduced; and task collaboration and system stability are ensured.

A multi-node, multi-container cluster system that generates, aggregates, and manages log files from services and components to be used for audit logs and to debug and perform other serviceability tasks provided by a vendor of the cluster system. Logs are collected from all components of the system and aggregated into a consistent format for user analysis and debugging. Embodiments provide a comprehensive way to parse and index vast numbers of log files that can then be packaged and displayed to a user in a way that facilitates analysis and debugging and / or efficient input to appropriate debugging programs.

The invention discloses a strategy fusion and differential game-based decentration cooperative guidance method, which comprises the following steps of: aiming at the problem of cluster system collapse caused by missed missile leading, combining the differential game with a consistency formation theory to enable missiles to approach a target in a predetermined formation so as to realize decentration; aiming at the problem that an HJI equation coupled by a nonlinear system is difficult to solve, constructing a neural network by utilizing a self-adaptive dynamic programming technology to carry out online solving on an optimal control strategy of pursuit and escape parties; for the dilemma that multiple escape strategies are difficult to select due to the fact that pursuit control strategies always appear in pairs, an interest value function is introduced, and a strategy fusion mechanism is provided; for the problem of a local minimum value possibly occurring in an artificial potential field method, a dynamic adjustment factor is introduced, and collision avoidance among intelligent agents of the whole cluster system is achieved.

The invention discloses an affine formation maneuvering control method for a heterogeneous cluster system and a related device, and relates to the technical field of cluster cooperative control, and the method comprises the steps: obtaining the state information estimation values of all leaders updated by a target follower at the current moment according to the communication relation of the target follower and a distributed observer, according to the state information of the target follower at the current moment and a preset affine formation maneuvering control protocol, control input of the target follower at the current moment is obtained through calculation; according to the state information of the target follower at the current moment, the control input of the target follower at the current moment and a preset dynamic model, position information of the target follower at the next moment is obtained through calculation; and according to the position information of the target follower at the next moment, an affine formation maneuvering control instruction at the current moment is determined, so that the expected formation configuration of the follower can be achieved when the control instruction is executed.

The invention discloses a cluster adaptive path planning method based on geometric PDE and PINN. The method comprises the following steps: firstly, modeling a dynamic model of a fault cluster with time-varying topology; secondly, through Riemannian manifold and gradient evolution based on heat flow, a fault cluster is constructed based on a parabolic PDE system, and a reference trajectory of a cluster system in an obstacle environment is obtained; and learning a steady-state solution of the cluster based on a parabolic PDE system by using a physical information neural network (PINN), and obtaining a fault-tolerant optimal control law of the cluster system based on geometric PDE and PINN under the conditions of time-varying topology, actuator fault and obstacle operation. According to the method, it is ensured that the intelligent agent navigates to the target position from the initial position along the dynamic reference trajectory, meanwhile, endogenous compensation of the fault effect is achieved, and unified solution of fault compensation and motion planning is achieved.

The invention discloses a GPU server cluster system and a GPU scheduling method, and relates to the field of data processing. The system comprises a plurality of GPU service nodes, a data transmission unit, a task segmentation unit, a GPU scheduling unit and a data processing unit. The data transmission unit is used for acquiring a to-be-processed video image data set; the task segmentation unit is used for generating a video image processing task and segmenting a plurality of subtasks; the GPU scheduling unit is used for determining load performance scores according to the load performance parameters, sorting the GPU service nodes from high to low based on the load performance scores, determining a target GPU node and distributing subtasks; the target GPU node is used for executing the subtask to obtain processed partial video image data; and the data processing unit is used for merging to obtain a processed video image data set, sending the processed video image data set to the data demand end, and sending video image processing task completion information to the GPU scheduling unit. The GPU computing power can be effectively scheduled, and the performance of the GPU server cluster is improved.

Systems and methods described herein relate to the assignment of user devices to workload clusters. Resource utilization on a plurality of user devices is monitored. A device agent on each user device may be used to monitor the resource utilization. A workload execution request identifies resource requirements of a cluster workload. The workload execution request is assigned to a cluster based on the resource requirements of the cluster workload and the resource utilization on the plurality of user devices. The cluster comprises the plurality of user devices. The workload execution request is caused to be executed on the cluster. Each user device executes user workloads and a respective portion of the cluster workload.

Providing issue resolution in a cluster system by monitoring system operation to detect occurrence of a system error, and automatically generating, upon detection of the error condition, a core file for a user node. The core file captures a current memory state of a respective node, where the current memory state comprises system statistics, system information, and logs. An intelligent core debugger extracts information from a core file to generate a core file report that is sent to a vendor for a quick determination of whether sufficient information is in the report to allow the vendor to recommend a fix, or whether further information from is required, including the core file itself, if necessary. This prevents the need to send an entire core file to a vendor in every instance of a system fault.

Methods and systems for generating federated learning models. In some aspects, the system receives, from each client device, user data profiles that are anonymized with respect to users associated with user data stored locally at a client device. The system processes the user data profiles to generate a plurality of clusters. For each cluster, the system transmits, to one or more client devices corresponding to a cluster, a first instruction to train a machine learning model on user data corresponding to user data profiles included in the cluster and a second instruction to validate the machine learning model with respect to user data corresponding to one or more clusters of the plurality of clusters other than the cluster to generate a prediction accuracy metric. The system determines, from the plurality of clusters, a first cluster based on associated prediction accuracy metrics.

A method includes receiving a dependency chain matrix from each of a plurality of service end points in a multi-cluster system converging the received dependency chain matrices into a converged data matrix, sharing the converged data matrix as a side car module to each of the plurality of service end points, receiving a service request by an external user or service for at least one of the plurality of service end points, validating an authorization for the service request based on the converged data matrix, and generating a label that is appended to the service request and passed downstream to the plurality of service end points.

A multi-node, multi-container cluster system that generates, aggregates, and manages log files from services and components includes a serviceability suite of functions for the system as it performs its tasks of supporting applications for a user. A logging function records events and states of nodes as the applications are run, an audit log processing function processes and compiles the logs for audit purposes, and the relevant log data for is provided in the form of uniform core files. Failure processing functions and debugging tools are provided to help analyze, debug, and correct issues that may be encountered. A support bundle processing function is provided to generate and process the log, debugging, failure, and other information for transfer to vendors or other third parties tasked with supporting the cluster network.

A power distribution network elasticity improvement method based on distributed new energy aggregation relates to the technical field of power systems, and comprises the following steps: S1, establishing a device-cluster-system three-level collaborative architecture, and collecting operation data of distributed energy nodes in real time through digital twin dynamic modeling; s2, dynamically adjusting a virtual power plant aggregation boundary based on a deep reinforcement learning algorithm, and optimizing a resource response strategy; s3, a multi-time-scale optimization framework is adopted, and space-time cooperative elastic enhancement is achieved in the pre-disaster stage, the in-disaster stage and the post-disaster stage; s4, realizing credible evidence storage and compensation liquidation of the resource scheduling instruction through the block chain; aiming at the problems of dynamic response delay and insufficient space-time dimension collaboration in the existing distributed energy aggregation technology, the invention realizes full-chain elastic improvement from minute-level dynamic aggregation to hour-level recovery optimization by constructing a three-level collaboration architecture, and solves the problems of low resource utilization rate and slow after-disaster recovery speed in the traditional method.

The invention discloses a unified representation and dynamic knowledge reasoning method and device for an intelligent cluster system, and the method comprises the steps: determining a decyclization task sub-graph based on a composite task state space model of the intelligent cluster system through the priori information; constructing a real-time data sub-graph by using a real-time data feedback mechanism, and constructing a traceability mode sub-graph based on the decyclization task sub-graph; candidate traceability sub-graphs with the highest matching similarity with the traceability mode sub-graphs are determined from the real-time data sub-graphs to serve as traceability sub-graphs; constructing a reasoning path set based on the decyclization task subgraph, and calculating a probability value of each reasoning path in the reasoning path set by using a path probability model constructed based on a Bayesian conditional probability chain rule; and screening out the reasoning paths of which the probability values are smaller than a preset probability value in the reasoning path set by utilizing a dual path pruning strategy, and adding the reasoning paths containing abnormal nodes in the reasoning path set into an alarm path set so as to provide support for state prediction and risk management of the intelligent cluster system.

The invention discloses a system quick recovery method and system based on system mirror image management, and the method comprises the steps: initializing a boot partition of a functional blade in a heterogeneous cluster system and a boot flag bit of the boot partition, the boot flag bit being used for BIOS reading to select a corresponding boot partition when the functional blade is started; the method comprises the following steps: triggering a job operating system or a boot operating system of a target function blade to restart through a soft shutdown interface, and modifying a start boot flag bit in a BMC (Baseboard Management Controller) of the target function blade, so that the target function blade is sequentially restarted and switched into the boot operating system from the job operating system; and executing system quick recovery or backup for the target system partition under the boot operating system, and then restarting and switching to run under the specified job operating system so as to complete system quick recovery or backup. According to the method, the characteristics of the heterogeneous cluster system are combined, and automatic and flexible system mirror image fast rollback recovery is achieved when the heterogeneous cluster system has a serious fault.

The invention relates to the technical field of cooperative computing, in particular to a distributed peer-to-peer multi-container cluster system which comprises a basic resource layer composed of basic hardware resources of multiple nodes, and the basic hardware resources comprise a computing server, a storage server, network equipment, safety equipment and intelligent acceleration hardware; the operating system layer is composed of maneuvering edges corresponding to a plurality of nodes and provides resource and service operation support for the container application load; the cloud management layer is composed of a plurality of container cloud management platforms and is distributed and operated on each node according to a high-availability deployment architecture to realize unified collaborative scheduling of cluster resources and applications; the data synchronization module is used for ensuring the consistency of metadata among the container cloud management platforms by adopting a distributed consistency protocol; and the application scheduling module receives the application deployment request, and issues the target application and the distribution strategy to the corresponding target node or node cluster to complete the scheduling of the target application.

The invention discloses a combined fault monitoring method and device based on a large model, a medium and equipment. The method comprises the following steps: collecting and preprocessing multi-dimensional data; extracting multi-source feature data based on a convolutional neural network and a long-short term memory network; performing redundancy elimination and fusion on the multi-source feature data; and cluster collaborative fault modeling is carried out, and global health assessment is carried out. According to the method, a deep learning large model is established through a convolutional neural network and a long-short term memory network, and effective fault features are automatically extracted from high-dimensional heterogeneous sensor data; through feature fusion of multi-source feature data, dimensionality reduction and fusion are carried out on features extracted from the sensor data, redundant information is removed, and a unified feature vector is generated; through the cluster collaborative fault modeling method, the identification problem of an equipment fault propagation cluster-level modeling path is solved, the globality and comprehensiveness of a fault monitoring system are improved, and fault monitoring of a cluster system is achieved.

The embodiment of the invention provides a cluster collaborative navigation method, a control system and a storage medium. The method comprises but is not limited to the technical field of navigation. The method comprises the following steps: in an upper layer module of a gene regulation and control network, determining a form boundary curve according to first position information of first execution equipment, second position information of second execution equipment and third position information of an obstacle; in a lower layer module of the gene regulation and control network, determining a tangential propulsion speed component and an offset correction speed component according to the form boundary curve and the first position information; and in a lower layer module of the gene regulation and control network, according to a visual adjacent distance regulation speed component, a tangential propulsion speed component and an offset correction speed component of the visual projection field, determining a target linear speed and an angle parameter. According to the embodiment of the invention, collaborative navigation and dynamic form maintenance of the cluster system can be realized.

Systems and methods are provided for validated transfer of replaceable hardware components within a computing cluster comprised of multiple IHSs (Information Handling Systems), such as rack-mounted servers. During factory provisioning of the IHS, a factory-signed inventory certificate is uploaded to each IHS that identifies factory-installed hardware of the IHS. This inventory certificate is used to validate that the detected hardware of the IHS is genuine factory-installed hardware. Upon a successful validation of detected hardware by an IHS, a cluster management console is notified. Based on the notifications of validated hardware by the IHS, the cluster management console identifies replaceable hardware components installed within IHSs that are members of the computing cluster. The cluster management console generates a delta certificate that authorizes operation by the IHS of the replaceable hardware components installed within IHSs that are members of the computing cluster.

Techniques for limiting or reducing storage accessibility “hiccups” in active-active clustered systems that perform large writes. The techniques can include executing, by a surviving, failover, or rebooted node of an active-active clustered system, a specialized recovery protocol that includes treating each large write request from a host computer as a plurality of small write requests while execution of the specialized recovery protocol is in progress, draining all dedicated sub-ubers for a primary and secondary node of the active-active clustered system, and, having completed execution of the specialized recovery protocol, resuming normal treatment of large write requests from the host computer. In this way, the need and complexity of managing large writes and maintaining their corresponding sub-uber information during recovery from a forced reboot, crash, or disaster involving the primary or secondary node can be avoided, and storage accessibility hiccups due to performing the large writes can be limited or reduced.

The invention relates to the technical field of computers, and discloses a server state monitoring method and device and a server cluster system, and the method comprises the steps: obtaining a reference file of a to-be-tested server; periodically scanning the hardware devices in the server to be tested to obtain current state information of each hardware device; generating a current state file of the to-be-tested server according to the current state information of each hardware device; comparing the difference between the reference file and the current state file of the to-be-tested server to obtain state change information of the to-be-tested server; and generating a state monitoring log of the to-be-tested server according to the state change information of the to-be-tested server. According to the method provided by the scheme, the hardware equipment state information of the to-be-tested server is automatically scanned regularly, and the state monitoring result of the to-be-tested server is determined by comparing the reference file, so that the state monitoring efficiency and timeliness of the server are improved.

The invention provides an event-driven privacy protection method for cooperative formation of unmanned aerial vehicle cluster systems, and relates to the technical field of unmanned aerial vehicle cluster control. Comprising the following steps: modeling a communication topology of an unmanned aerial vehicle cluster by using a graph theory, and determining a formation center of a directed strong connectivity balance topology; establishing a motion model of the unmanned aerial vehicle cluster according to the topology of the unmanned aerial vehicle cluster and the formation center; on the basis of a traditional privacy protection collaborative algorithm, a control item based on an event triggering strategy is introduced, real communication data are protected by utilizing a finite time mask function, and a novel privacy protection composite control law is provided; by using the design, the cooperative formation of the unmanned aerial vehicle cluster system is realized, and the privacy of the initial position and speed of each unmanned aerial vehicle is ensured. According to the invention, the problem of network topology limitation in the unmanned aerial vehicle privacy protection formation realized by using deterministic interference in the prior art is solved.

Systems and methods described herein relate to the assignment of user devices to workload clusters. Resource utilization on a plurality of user devices is monitored. A device agent on each user device may be used to monitor the resource utilization. A workload execution request identifies resource requirements of a cluster workload. The workload execution request is assigned to a cluster based on the resource requirements of the cluster workload and the resource utilization on the plurality of user devices. The cluster comprises the plurality of user devices. The workload execution request is caused to be executed on the cluster. Each user device executes user workloads and a respective portion of the cluster workload.

The invention relates to the technical field of power distribution and power saving of a power system, in particular to an AI cluster system-based CPU (central processing unit) power distribution method for a PC (personal computer), which comprises the following steps of: step 1, collecting long-short period data, step 2, analyzing long-period and short-period habits, step 3, dynamically adjusting habit weights, step 4, detecting and responding to abnormity, and step 5, distributing power to the PC. Continuously optimizing and updating; according to the scheme, long-period and short-period data are comprehensively collected and finely analyzed, so that the accuracy and the dynamic performance of CPU power distribution are realized. Long-period data can capture long-term trends, short-period data can reflect short-term changes, strategies are adjusted in time to adapt to dynamic requirements, time sequence analysis, clustering algorithms and the like are used for deep mining and classification, a simulation environment is constructed to determine an initial scheme, CPU operation and energy utilization efficiency is improved in the mode, different user scene requirements are met, and user experience is improved. And smooth, stable and efficient use experience is provided.

The invention relates to the technical field of distributed synthetic aperture radar (SAR) imaging detection and cluster guidance, in particular to a mobile platform bistatic SAR moving target imaging and positioning parameter estimation method and system, through the method provided by the invention, distance migration correction of a moving target is realized by adopting distance frequency inversion transformation, and the positioning accuracy of the moving target is improved. According to the method, an accelerated multi-component QFM signal parameter estimation method is optimized by adopting a genetic algorithm in azimuth, and finally, state parameter estimation on a maneuvering target with acceleration is realized by adopting a combined aperture method based on each order of estimated Doppler parameters, so that the condition of failure in the prior art is avoided. According to the method, imaging and positioning of the maneuvering target by the maneuvering platform bistatic SAR are achieved, the application range of the current bistatic SAR is expanded, and the method mainly relates to point multiplication and fast Fourier transform (FFT) and is high in operation efficiency, so that the method is suitable for the scene that terminal guidance application is conducted on the moving target by a cluster system.

The invention discloses a domestic VPX server-based cluster system and an application service VIP attachment method, and relates to the field of domestic information systems. The system comprises three computing modules, each computing module adopts a domestic processor, each computing module is provided with a Nomad server and a Consul server, the computing modules are converted into a management node mode, and a one-master two-slave architecture is formed through a master selection mechanism; a Nomad client and a Consul client are respectively operated on each computing module, and the computing modules are converted into computing node modes, so that the computing modules have both management and computing functions. During work, a Nomad cluster is constructed based on a domestic VPX server, mixed arrangement and full-life-cycle management of containerized application services and traditional application services are achieved, resource management and high service availability are supported, and reliable and stable operation of an information system is guaranteed; meanwhile, a VIP function is provided for the application service, and transparent and non-inductive access of an external client or the application service is supported when the application service is migrated.

A computing cluster system, a security authentication method, a node device and a storage medium are provided. A service grid system for implementing an identity-based identification authorization mechanism between at least two services is added in the computing cluster system. In the system, an identity registration component reports, to a management and control component, acquired diversified identity data corresponding to a first service; according to the diversified identity data, the management and control component generates a security policy and a digital certificate, which correspond to the first service, and issues same to a corresponding authorization proxy component. The authorization proxy component performs identity verification for the corresponding service according to the security policy and the digital certificate, and performs communication for the at least two services by proxy after the identity verification is passed.

The invention provides a task scheduling method and device based on Kubernetes, the method is applied to a controller, and the method comprises the following steps: receiving a task submission request; calling a user-defined task segmentation program interface, and splitting the received task into a plurality of subtasks through a task segmentation program; selecting a plurality of target nodes matched with the task from the nodes based on the obtained resource state of each node of the Kubernetes; scheduling the plurality of sub-tasks to the corresponding target nodes, so that each target node processes the received sub-tasks through a task processing program by calling a task processing program interface; and calling a user-defined result merging program interface, obtaining processing results of the subtasks by the target nodes through a result merging program, and merging the processing results to obtain a final processing result. According to the task scheduling method based on Kubernetes, the node resource state can be dynamically sensed, fine-grained segmentation and combination of tasks can be achieved, and the task processing efficiency can be improved.

The invention relates to the technical field of subway communication, and discloses an SIP (Session Initiation Protocol)-based special wireless cluster interconnection and intercommunication system for a subway network, which comprises a hardware component and a software component, the hardware component comprises an SIP (Session Initiation Protocol) server, an SIP scheduling terminal, an Internet gateway and a TETRA (Terrestrial Transport Radio Access) system or a broadband cluster The software assembly comprises an SIP protocol stack, signaling conversion and interaction software and management and configuration software. According to the special wireless cluster interconnection and intercommunication system for the subway network based on the SIP protocol, by introducing the SIP standard protocol, the technical barrier is successfully broken through, comprehensive interconnection and intercommunication of multiple brands of TETRA and broadband cluster systems are achieved, a fixed station does not need to be used for transferring, the communication efficiency and quality are remarkably improved, meanwhile, the system keeps the structural integrity of an original system, and the service life of the system is prolonged. The method simplifies the configuration process, reduces the cost, has strong expandability, and supports the seamless access of a new line in the future.