119 results about "Distributed management" patented technology

The invention discloses a large model data distributed management method and device, equipment and a storage medium, and relates to the technical field of data processing, and the method comprises the steps: segmenting large model training data to obtain a plurality of data blocks, and determining a predicted access frequency based on a long short-term memory network model and the historical access frequency of the data blocks; caching the large model training data corresponding to the data block to a corresponding data cache layer by utilizing the predicted access frequency; constructing a resource portrait by using the static attribute and the dynamic index of the GPU node, and allocating the large model training task to a target GPU node by using a preset hybrid strategy and the resource portrait based on the predicted access frequency and the storage position corresponding to the data block in the data cache layer; when it is monitored that the large model training task on the target GPU node is executed, periodic snapshot is conducted on the large model training task through a distributed snapshot algorithm, and the obtained complete data state is stored in a distributed storage center. And the resource vacancy rate is reduced.

A computer-implemented method of distributed management of locks includes: receiving, by a plurality of servers, a plurality of concurrent requests; performing lock preemption on the distributed lock using a synchronization mechanism; only one server uses a synchronization mechanism to obtain the distributed lock; updating the lock value of the distributed lock by only one server; and executing the business logic. Lock preemption is performed on the distributed lock by the respective server, including comparing a current memory value of the distributed lock to an expected value. The method further comprises the steps that if the current memory value is matched with the expected value, only one server modifies the current memory value into an updated value; writing the updated value into a memory of the distributed lock; and if the execution of the service logic fails, recovering the memory value of the distributed lock to the current memory value.

The invention relates to the field of software development application data processing, and discloses a software development application data processing method based on an AI large model, which comprises the following steps: deploying an anomaly detection model in a data access layer, and carrying out dynamic anomaly detection and real-time cleaning on software development application data, based on historical security event feedback, optimizing the scoring model and grading the software development application data; anti-association desensitization training is carried out on the software development application data, homomorphic encryption is carried out, association attack simulation testing is carried out regularly, and parameters are adjusted; carrying out distributed management on the key, and carrying out audit tracking based on a block chain; accessing a regulation database based on a dynamic compliance engine, analyzing a regulation to generate a rule base, deploying check points in a data key link for real-time detection and blocking, and performing secure erasure verification; dynamic scheduling resources are predicted in combination with a large model, and the cost is monitored in real time through a monitoring system; nodes are deployed on multiple platforms, an aggregation parameter updating model is encrypted, and vulnerabilities are scanned regularly.

The invention discloses a multi-time-scale coordination control method and system for a virtual power plant with new energy access, and aims to solve the challenge of the fluctuation of new energy on the stability of a power grid. The method comprises the following steps: in a day-ahead scheduling stage, constructing a baseline scheduling plan based on natural conditions and historical loads, and allocating energy storage node demands through a block chain; in the intra-day rolling optimization stage, the plan is dynamically adjusted through short-term prediction, and energy storage requirements are redistributed; in the construction stage, a digital twinborn technology is adopted to simulate dynamic interaction between new energy and bus nodes and energy storage nodes; in the real-time control stage, energy storage demand sharing is realized in combination with model prediction control and a block chain. The system comprises a baseline scheduling module, an intraday optimization module, a virtual power plant construction module and a real-time control module. According to the method, through multi-time scale cooperation, block chain distributed management and total element simulation, the power balance capability is remarkably improved, the adaptability to the uncertainty of natural conditions is enhanced, and the operation efficiency and transparency are improved.

The invention discloses a computer self-check cycle optimization method based on the Internet of Things in the technical field of server maintenance, and relates to the technical field of server maintenance. Server node operation state data is collected in real time, a dynamic load evaluation model is constructed, and a comprehensive load index is calculated; and dynamically dividing performance states based on the threshold values L1 and L2, and adjusting a self-checking period and a module priority. Calculating a single-node risk index Ri by adopting a fault risk model, generating a fault generalization risk factor R1 through a probability graph model in combination with network topology, and triggering a linkage mechanism; dividing and dynamically adjusting a self-checking period according to the early warning level; during advanced early warning, redundant resources are distributed to preferentially process key faults, edge calculation is combined to carry out data cleaning and standardization, distributed management is realized through a consistent Hash algorithm, and finally, a visual interface is used for displaying an optimized scheduling result, so that a closed-loop optimization system of data acquisition, risk prediction and resource scheduling is formed.

The present invention relates to the field of software development application data processing, and discloses a software development application data processing method based on an AI large model, including: deploying an anomaly detection model at the data access layer to perform dynamic anomaly detection and real-time cleaning on software development application data, optimizing the scoring model based on historical security event feedback and classifying the software development application data; performing anti-correlation desensitization training on the software development application data, performing homomorphic encryption, regularly conducting association attack simulation tests and adjusting parameters; managing keys in a distributed manner and performing audit tracking based on the blockchain; accessing a regulation database based on a dynamic compliance engine, parsing regulations to generate a rule library, deploying checkpoints at key data links for real-time detection and blocking, and performing secure erasure verification; dynamically scheduling resources in combination with large model prediction, and monitoring costs in real time through a monitoring system; deploying nodes on multiple platforms, encrypting and aggregating parameter update models, and regularly scanning for vulnerabilities.

The invention discloses a customer service work order management system based on network employer service, which relates to the technical field of work order management and comprises a multifunctional service module, a distributed management module, an elastic module, an operation and maintenance monitoring module and a real-time notification module. The multifunctional service module comprises a user service unit, a work order service unit, a communication service unit, a knowledge base service unit and a data analysis unit, and is used for serving a network employer; the distributed management module is used for verifying user validity, determining user legality and creating a work order based on a legal user; and the elastic module is used when the customer service work order service response is overtime. According to the invention, omnibearing real-time monitoring and deep analysis are carried out on the system, discovered problems and work order state change conditions are pushed to related personnel in time, timely transmission of information is ensured, failure retry is carried out based on an exponential backoff algorithm, the reliability of notification is ensured, and notification loss caused by network fluctuation is avoided.

The invention relates to the field of energy storage battery communication management, and particularly discloses a large-scale multi-node high-throughput data communication system of an energy storage battery, which comprises an equipment layer, a communication layer and a management scheduling layer, the management of the equipment layer on the battery comprises an IP mode and grouping management based on a battery state; a broadband industrial bus of the communication layer is used for data transmission among a BMU, a BCU and a BAU in the BMS and data interaction between the BMS and an EMS; the distributed domain bus is used for dividing the BMS into different domains, and cluster communication between BMS nodes is achieved in each domain. Cluster management is realized by using a distributed domain bus, the problems of complex wiring and difficult maintenance of traditional networking are solved, unified scheduling of energy storage devices of different manufacturers is realized through a self-defined domain address strategy, and addressing efficiency and distributed management and control capability of large-scale nodes are optimized.

The invention relates to the technical field of image processing and data management, and discloses an infrared image data classification filing method and platform, and the method comprises the following steps: 1, collecting infrared image data from a plurality of different types of infrared sensors, the sensors including an infrared thermal imaging camera, a visible light camera and a radar; 2, the collected infrared image data are preprocessed, noise removal, temperature correction and image enhancement are included, and wavelet transform and a deep convolutional neural network are adopted for noise removal; through intelligent classification and management, the infrared image data processing efficiency is improved, manual intervention is reduced, the error rate is reduced, the data storage and retrieval speed is optimized by adopting multi-level storage and distributed management, the storage cost is reduced, the data safety and reliability are guaranteed through redundant backup and data de-duplication, and the data storage and retrieval efficiency is improved. And long-time reliable storage and efficient recovery of the data are ensured.

The invention discloses a water conservancy informatization management system based on a digital twinborn model, and relates to the technical field of water conservancy management, and the system comprises a block management module which is based on the equipment, terrain and hydrological information data of water conservancy facilities, management blocks are divided according to the equipment, terrain and hydrological information data characteristics of the water conservancy facilities, computing equipment is locally deployed in each management block, and the local computing equipment of the management blocks is utilized to distribute management computing tasks according to the load capacity. Management blocks of equipment, terrains and hydrological environments of water conservancy facilities are divided into a plurality of gridding node units through the block management module, data storage and calculation tasks are independently undertaken through independent monitoring management of each management block, the data load pressure of a single central server is reduced, and the management efficiency of the water conservancy facilities is improved. Response delay and fault risks caused by overload of the central server are avoided, and stable operation of the water conservancy informatization management system is guaranteed.

The invention provides a heterogeneous GPU resource pooling method and system combining Ray and Volcano, and relates to the technical field of GPU resource scheduling. According to the method and the system, firstly, a Ray + Volcano framework is constructed on a Kubernetes cluster; then deep learning model training is carried out based on a Ray + Volcano framework, distributed management is carried out on deep learning training tasks through a task scheduling framework of a Ray cluster, and task load balance is ensured; and a resource management mechanism of Volcano job scheduling is utilized to realize control and optimal distribution of GPU resources. According to the method and the system, rapid execution of large-scale tasks and efficient utilization of resources are realized through the efficient distributed computing capability of Ray and the dynamic resource scheduling and isolation mechanism of Volcano.

The invention discloses a cloud platform design method based on a Docker container technology, and relates to the field of cloud computers. The method comprises the following steps: deploying a plurality of containerized target application services in a distributed management cluster, and respectively determining the number of container instances required by the plurality of target application services; for each target application service, distributing a plurality of container instances to different working nodes for operation according to a scheduling strategy and the number of the container instances; configuring a multi-level network isolation strategy for the plurality of target application services; constructing a label-based dynamic container instance scheduling strategy, and performing real-time scheduling on the container instances of the plurality of target application services through the dynamic container instance scheduling strategy; and copying the cluster management state among the plurality of management nodes in real time, and carrying out synchronization and regular backup on the application state of the running container instance, so that the running states of all nodes in the distributed management cluster are consistent. The method can improve the reliability of cloud platform design.

The disclosed system provides distributed management and control of satellite constellations. Such distributed management and control that involves a plurality of satellites, a plurality of ground stations, a mission control station, and a data center that hosts a portion of an Internet cloud. The system resolves issues that have plagued legacy attempts. Specifically the disclosed system includes optimized communication with satellites that leads to more efficient transfer of data and payloads to or from or by or between satellites and ground control systems. This is achieved through use of micro-batching of communications. In this manner, data may be processed and more effectively transmitted to and from more than one satellite and back to one or more ground stations, including coordinated transmissions of portions of a micro-batch between two or more ground stations.

The invention discloses a rule engine and low-code dual-drive power grid digital graph fusion treatment platform and method. The platform comprises an infrastructure layer, a data layer, an algorithm engine layer and a business application layer. And the data layer is used for accessing and managing multi-source power grid data. And the algorithm engine layer comprises a rule engine module for configuring and executing data verification, digital graph association and exception handling rules, a data quality diagnosis algorithm module for performing data quality diagnosis to identify exceptions, and a digital graph fusion driving engine module for realizing bidirectional linkage between data and graphs. And the business application layer comprises a low-code arrangement platform for customizing a governance scene page and a work order process, a digital graph integrated display module for displaying multi-source power grid data on the same screen, and a closed-loop work order management module for automatically generating and distributing governance work orders. The method constructs a depth digital graph fusion system, improves the anomaly positioning and treatment efficiency, achieves automatic management and control and whole-course tracing, does not need a large amount of code development, and shortens the response period of a new business scene.

The invention discloses a cloud edge efficient synchronization mechanism and distributed management scheduling system based on crop water shortage data, and relates to the technical field of intelligent agriculture. The system comprises an edge end, an edge computing node module, a cloud edge communication module and a cloud storage and management module, wherein the edge end is used for collecting multi-source original data related to crop water shortage; the edge computing node module is used for preprocessing the multi-source original data and extracting a feature vector comprehensively reflecting a crop water shortage state; the cloud side communication module is used for constructing a two-way communication link and adjusting the data transmission rate according to an adaptive window adjustment strategy; and the cloud storage and management module is used for carrying out encrypted storage on the data by adopting a distributed hash table technology and carrying out dynamic scheduling and distribution on the stored data by utilizing a distributed management scheduling algorithm. According to the invention, efficient collection, accurate transmission, intelligent scheduling and safe storage of crop water shortage data can be realized, and agricultural data management precision and decision-making efficiency are improved.

The invention discloses a distributed remote multi-available-area computing power scheduling method, and relates to the technical field of distributed management, and the method comprises the following steps: a management node collects available area state information and reports the available area state information to a main service node, so as to construct a dynamic global resource view; after receiving the calculation task, the main service node analyzes and splits the calculation task into sub-tasks; calculating the cost of scheduling to different available regions for each sub-task by constructing an adaptive cost function, and selecting the sub-task with the lowest cost as a target available region; distributing the sub-tasks to management nodes of the target available area, and distributing the sub-tasks to computing nodes for execution by the management nodes; the result is aggregated and returned to the main service node; meanwhile, the state of the available area is monitored, and tasks in the fault available area are rescheduled. Through combination of centralized coordination and distributed execution, computing power scheduling with low delay, high availability and high resource utilization rate is realized.

This application relates to the field of device connectivity technology and discloses a method and system for managing the connection of multiple cameras based on portable WiFi. The method includes: calculating the resource constraint coefficient in the portable WiFi environment; generating a distributed management configuration table for the management tasks of multiple cameras in the portable WiFi based on the resource constraint coefficient, and performing alternating active scanning and passive listening to obtain a camera access priority table; allocating resources to the camera devices in the camera access priority table to obtain a multi-camera resource allocation strategy, and dividing and allocating three-layer transmission channels based on the multi-camera resource allocation strategy to obtain a multi-channel data transmission control table; establishing a multi-level resource feedback mechanism in the portable WiFi environment based on the multi-channel data transmission control table, and dynamically calculating the bandwidth allocation ratio between camera devices to obtain a collaborative transmission optimization scheme, thereby improving the utilization efficiency of limited bandwidth resources and realizing the efficient collaborative work of multiple camera devices.

An integrated circuit (IC) device includes a functional circuit and a distributed management circuit that includes a plurality of configuration interface manager (CIM) circuits that receive respective programming partitions as configuration packets over a first communication channel (e.g., a network-on-chip or NoC), the configuration packets being configured to communicate with the functional circuit. And performing management operations on the respective regions of the functional circuit in parallel with each other based on the respective configuration packets, the management operations including providing configuration parameters to the respective regions of the functional circuit. The configuration packets may be streamed from the central manager to the CIM circuitry and / or read by a direct memory access (DMA) engine of the CIM circuitry. The central manager may configure the CIM circuitry and the NoC over a second communication channel (e.g., a global communication ring interconnect) during the initialization phase. The CIM circuitry may include respective packet processors, random access memories, authentication circuitry, error detection circuitry, and interconnect circuitry having a standardized bus width.

The application discloses a kind of distributed data encryption sharing methods based on block chain, and using block chain realizes the traceable management of search and authorization process.The scheme is first by multi-authorization center to carry out distributed management to attribute domain, and generates attribute key associated with identity for user;In access verification link, construct parameter masking mechanism based on bilinear pairing, so that user attribute private key is not directly exposed in block chain verification and query process.Secondly, data owner carries out layered encryption to shared data and session key, and constructs the improved multi-key search structure, index and ciphertext positioning information are recorded to block chain, to support fast positioning target ciphertext.In addition, the application uses linear secret sharing to set access structure, and designs subset determination mechanism of general attribute name set, binds the download permission of ciphertext with the attribute condition that user can satisfy, to prevent overreach download behavior.Finally, with the help of cloud, outsourcing decryption is executed to generate intermediate result, and terminal only needs to complete light recovery and decryption.The application can give consideration to fine-grained control, high-performance search access and terminal side low load, and is suitable for secure data encryption sharing in resource-limited scene.

The invention provides a data processing method, device and equipment, a medium and a program product, and relates to the technical field of cloud computing big data edge computation.The method comprises the steps that multi-modal data subjected to data preprocessing is obtained and stored in a data lake and warehouse integrated system, and the data lake and warehouse integrated system is used for storing the multi-modal data; wherein the data lake and warehouse integrated system is obtained by integrating a data warehouse and a data lake cluster; extracting modal features from the multi-modal data based on a deep learning technology, and generating feature vectors of the modal features; the feature vectors are mapped to an equal-dimensional space, the similarity of all modal features is calculated, a comparative learning model is constructed based on the minimum similarity of all the modal features, and the comparative learning model is used for conducting distributed management on the multi-modal data stored in the data lake and warehouse integrated system. According to the heterogeneous data query method and device, the comparative learning model based on the modal features is constructed through the deep learning technology, distributed management of the data is carried out according to the result of the comparative learning model, and efficient heterogeneous data query is achieved.

The invention discloses an external open interface management method and device, a storage medium and a computer program product, and relates to the technical field of distributed management, and the method comprises the steps: obtaining an external access request through a unified request entrance; performing multi-dimensional validity check on the external access request; if the multi-dimensional validity check is passed, performing automatic authentication on the external access request, and obtaining authentication information based on the automatic authentication; and superposing the authentication information to the external access request to obtain a legal access request, and forwarding the legal access request to a corresponding target service in a routing manner. Through unified request admission, multi-dimensional validity check, automatic authentication and forwarding processing, refined and unified management of the external open interface is realized.

To provide a payment information management system and a payment information management method with which information on a payment made via a credit card is easily restored.SOLUTION: In a payment information management system 1, a payment terminal 10 obtains information on a credit card that has been used to make a payment, creates payment information according to the card information, sends the payment information and update request information for recording the payment information in ledger information, to a distributed management terminal, and stores the ledger information. The distributed management terminal 30 updates the stored ledger information according to the received update request information and payment information, and sends at least information on the update on the ledger information at least to a payment terminal 10 that differs from the payment terminal 10 via which the update request information has been sent.SELECTED DRAWING: Figure 1

The present invention discloses a method for analyzing electric power big data for distributed management of smart grids, which relates to the technical field of power grid data management, including: recording all electric power facilities connected to the smart grid distributed management system as target electric power facilities; selecting at least one master management electric power facility, and recording the target electric power facilities other than the master management electric power facility as slave management electric power facilities; dividing all electric power data of the smart grid distributed management system into low-frequency call electric power data and high-frequency call electric power data; obtaining electric power data request instructions of the target electric power facilities in real time, and classifying the instructions, and executing and responding to all electric power data request instructions through the master management electric power facility and the slave management electric power facility respectively. The advantages of the present invention are: realizing the read-write separation cluster management of smart grid data, while ensuring that the read-write separation cluster during the operation of the power grid is consistent with the actual operation status of the power grid, thereby achieving high reliability of the system.

Customers of a software platform, such as a unified communications as a service platform, are enabled to control their own encryption keys used to encrypt and decrypt data from various communication services in the software platform. A key broker server is employed to map encryption and decryption requests from servers in the platform to key management servers of customers based on user identifiers. Examples of data encrypted may include conference recordings, webinar recordings, phone call recordings, voicemails, emails, and calendar tokens.

A kind of distributed attack-resistant domain management and control method and system for giant constellation, the present application relates to satellite constellation management and control technical field, the present application aims at solving the problem of cooperative management and control and attack-resistant resilience maintenance of multi-layer heterogeneous giant constellation in dynamic topology environment.Technical points: the present application takes the management and control node satellite as core, combines with the cluster head satellite and ordinary satellite in domain, forms distributed management and control domain structure.The structure undertakes cross-orbit communication switching and distributed cooperative management and control function, effectively deals with the cooperative management and control challenge brought by the difference of multi-orbit satellite in operation cycle, communication topology and communication time delay;Introduce multi-element maximum entropy trust model: based on fuzzy C clustering algorithm and maximum entropy principle, construct dynamic trust evaluation system, and comprehensively quantitatively analyze the trust degree of distributed nodes.Through real-time identification of high-risk nodes and implementation of isolation measures, significantly reduce the interference of malicious satellite to inter-satellite data transmission, improve system security.The present application realizes the efficient cooperative management and control and attack-resistant ability enhancement of multi-layer heterogeneous giant constellation, provides key technical support for stable operation of giant constellation under complex threat environment.

The invention discloses a node management method in a distributed management system, relates to the technical field of distributed systems, and aims to solve the problems of load scheduling rigidness, low fault positioning efficiency and poor cross-node collaboration consistency in existing node management. The method comprises a node state sensing module, a dynamic load scheduling module, an intelligent fault processing module and a consistency maintenance module, accurate distribution of node loads is realized through an improved weighted adaptive load algorithm, a dynamic dependency graph is constructed in combination with a global TraceID penetration technology and a graph convolutional network model, and rapid positioning and self-healing of fault root causes are realized. A lease mechanism is innovatively introduced to optimize node state consensus, and reliability and high efficiency of cross-node data synchronization are guaranteed through a hierarchical consistency protocol. The method breaks through the limitation of traditional static scheduling and manual troubleshooting, improves the adaptive capacity, fault tolerance performance and operation and maintenance efficiency of the system, is suitable for a large-scale complex distributed cluster scene, and has wide engineering application value.

The utility model discloses a kind of integrated energy demand side distributed management device, including case, mobile device, horizontal moving platform and detection box, partition installation space is equipped in case, with controller and power module are equipped;Mobile device contains moving wheel and support adjusting seat, support adjusting seat can be vertically moved to realize "mobile-fixed" state switching;Horizontal moving platform is driven support table horizontal movement by drive screw and drive hand wheel;Detection box is equipped on support table top, and multiple in-out interfaces are equipped on side wall.The device is flexible and stable support by mobile device, simplify interface docking by horizontal moving platform, rely on partition installation space and detection box to improve equipment integrated protection capability, and can be integrated photovoltaic inverter detection, energy consumption sensor adaptation function to realize intelligent management, effectively improve the real-time performance, security and operation efficiency of energy management.

A metadata synchronization method, apparatus, and system. The method is applied to a distributed file system. The system comprises a plurality of nodes, each node storing file system metadata, wherein a first node operates on a file and updates first metadata. The first node sends a first metadata update request to other nodes by means of a collision-free data synchronization protocol. The first node receives a second metadata update request sent by a second node on the basis of the collision-free data synchronization protocol, and the first node updates second metadata on the basis of the request. In the method, a distributed metadata management mode is used, thereby avoiding the problem in centralized management that a metadata management node is highly loaded and easily becomes a system bottleneck. The distributed management mode reduces the risk of a single point of failure, and facilitates system scalability and reliability. Metadata synchronization is performed on the basis of the collision-free data synchronization protocol, thereby ensuring data consistency, avoiding the problem of high delay caused by a complex election mechanism and a communication process, and improving synchronization efficiency.

The application provides a kind of intelligence calculation fusion network heterogeneous resource elasticity adaptation and deterministic scheduling system and method, belongs to industrial internet of things technical field, comprising: the system of establishing distributed management and centralized control under the scene of intelligence calculation fusion network;According to the distributed management and centralized control system, the two-stage random optimization problem of different time scales is established;For large-scale resource adaptation problem, a resource elasticity adaptation scheme based on heuristic algorithm is adopted;For small-scale scheduling sub-problems, a deterministic scheduling scheme based on deep reinforcement learning is adopted;Using the resource elasticity adaptation scheme and deterministic scheduling scheme, the optimal resource adaptation and scheduling result is obtained.The application can meet the resource flexible configuration demand under high dynamicity and uncertainty environment, ensure low delay and high reliability in data scheduling process.

A server includes physical compute nodes. Each physical compute node includes a host and physical management resources. The physical management resources include a physical management processor. The server includes a distributed hypervisor to provide a distributed application operating environment that is hosted by the physical management resources. The distributed hypervisor to allocate, from the physical management processors, virtual processors for the distributed application operating environment to execute applications to manage the physical compute nodes. The distributed hypervisor includes a plurality of hyper-kernels that are associated with respective physical compute nodes. Each hyper-kernel is hosted on the physical management resources of the associated physical compute node.