23 results about "Resource isolation" patented technology

The present disclosure provides a resilient (radio) access network ((R)AN) slicing framework encompassing a resource planning engine and distributed dynamic slice-aware scheduling modules at one or more network access nodes, edge compute nodes, or cloud computing service. The resilient (R)AN slicing framework includes resource planning and slice-aware scheduling, as well as signaling exchanges for provisioning resilient (R)AN slicing. The intelligent (R)AN slicing framework can realize resource isolation in a more efficient and agile manner than existing network slicing technologies. Other embodiments may be described.

Embodiments of the present disclosure provide a method and apparatus for resource isolation on a communication network. A method (400) performed by a first node for managing resources of a communication network comprises: receiving (S402), from a second node, a first request for a communication resource, the first request comprising: a first parameter indicating an isolation requirement for the communication resource; and transmitting (S404), to the second node, a first response indicating the communication resource that meets the isolation requirement of the communication resource. Exemplary embodiments of the present disclosure propose a mechanism that allows exchanging isolation and security requirements of communication resources in a standard way.

This invention relates to the field of communication management technology, specifically to a data traffic management method, electronic device, and storage medium. The cloud-based data traffic management method includes: receiving a function activation request for a target function and determining the remaining available resources corresponding to the target function; generating function activation parameters containing the remaining available resources and sending the function activation parameters to the device; wherein the function activation parameters are configured to: trigger the device to start the target function; accumulate resource consumption in real time during the operation of the target function; and perform resource control operations on the target function when the resource consumption reaches a first threshold determined based on the remaining available resources. This method stores the control logic and policies in the cloud, avoiding the risk of local firmware tampering; simultaneously, through function-level independent limits and real-time monitoring, high-traffic functions exceeding their limits only affect themselves, and status reporting is not interrupted, achieving precise resource isolation and control.

This application discloses a multi-tenant task scheduling method and related apparatus, relating to the field of cloud computing technology. It receives tasks to be scheduled and their metadata, determines dynamic priority information, and sends the task to a target priority queue if the tenant corresponding to the task has an effective quota. When a tenant has a concurrent consumption token in the target priority queue, it acquires the token, executes the task, and then releases the token. This application sets different dynamic priorities for tasks to be scheduled, ensuring that tasks are distributed to resource-isolated independent queues, guaranteeing priority processing of each task, and improving resource utilization. Before task publication, sliding window rate limiting is implemented, and tenant-level concurrent consumption tokens are distributed during task consumption, isolating tasks between tenants. In complex multi-tenant, high-concurrency production scenarios, this enables intelligent task adjustment, significantly improving resource utilization and scheduling efficiency.

This invention relates to the field of task scheduling technology, specifically to a multi-task processing method and system based on a large model, comprising the following steps: extracting resource parameters using a large model to generate a conflict intensity matrix, screening resource-overlapping node pairs, determining structural conflicts, constructing a dependency graph to calculate complexity scores and prune dense regions, extracting high-frequency paths to divide scheduling units, and generating a task priority sequence. In this invention, by constructing a conflict intensity matrix and combining node clustering features to identify task resource conflict relationships, the sensitivity and accuracy of task partitioning to resource distribution are improved. Nodes are divided into upper and lower bounds based on median deviation, and structural optimization and reorganization are achieved by combining resource type dissimilarity rates. Path depth and branch factors constitute the basis for complexity scoring. Weights are dynamically adjusted based on occupancy time to identify pruning dense regions. The most frequently called paths constitute the main path set. Non-main path tasks are divided according to node coverage relationships, enhancing the clarity of scheduling boundaries and resource isolation capabilities.

The application belongs to the technical field of AI development, and particularly relates to a localized one-stop AI development platform and method, which comprises at least one first control server, multiple computing power servers and a data center level Ethernet switch. The Ethernet switch and the multiple computing power servers are located at a computing interconnection layer. The at least one first control server is located at a data access layer and is a core management layer of a business network. The application adopts a hierarchical design network architecture to solve the technical problem of how to build an AI development platform with high performance, high reliability and resource isolation capability.

The application relates to a method, system, device and medium for realizing multi-tenancy based on directory partitioning. In the multi-tenancy mode, a first target root directory with a true tenant root directory attribute is created, and a target first-level subdirectory with a true target working directory attribute of a tenant is created under the first target root directory according to the tenant; the target first-level subdirectory is only visible and accessible to the corresponding tenant; a target second-level subdirectory with a true target namespace working directory attribute of a user under the corresponding tenant is created under the target first-level subdirectory, and the target second-level subdirectory is only visible and accessible to the corresponding user to realize resource isolation between different users; and the user can create a subdirectory with a true ordinary directory attribute under the corresponding target second-level subdirectory to realize file management index. The application realizes resource isolation between tenants and users by using directory partitioning of a file system directory, so as to meet the multi-tenancy and multi-user demand.

The application relates to the technical field of container resource isolation, and discloses a kind of intelligent fusion terminal micro application resource isolation method based on container technology, comprising: setting sampling period and continuously collecting the running behavior data of each container in each sampling period;Based on the running behavior data, the running correlation model between containers is constructed;Based on the running behavior data, identify abnormal running containers and identify the abnormal correlation path of each abnormal running container based on the running correlation model;Based on the abnormal correlation path, divide the minimum state set for each abnormal running container;Based on the minimum state set, respectively, each abnormal running container is isolated from resources.The application accurately cuts off the abnormal propagation path at the state level, realizes the controllable isolation of container anomaly and the stable operation of business continuity.

The application discloses a microcontroller-based Android container resource isolation and data security guarantee method, relates to the field of hardware-assisted embedded container security technology, and comprises the following steps: loading a special firmware to build a transparent isolation layer on a microcontroller; generating a dynamic resource configuration vector in response to an application startup request and writing the vector into a non-volatile register group, locking the physical access paths of processor cache channels, memory bandwidth and peripheral interfaces; splitting a data stream into transmission frames with encryption labels when cross-container data interaction occurs, and routing the transmission frames to target containers through a hardware bus arbitrator; monitoring the space-time correlation of container kernel instruction streams and peripheral level jumps, freezing the container clock domain when the security model is violated, and triggering a hardware interrupt to perform state snapshot forensics; and sending a specific pulse sequence to a memory particle to trigger capacitor active discharge and reset the processor branch predictor history record when the container is terminated. The application realizes the resource exclusivity, data security and threat defense capability of the Android container.

This application relates to a method, apparatus, electronic device, and storage medium for generating materialized views. The method involves: filtering target query patterns from database query logs; generating candidate materialized view creation statements corresponding to the target query patterns; creating candidate materialized views based on these statements in a resource-isolated sandbox environment and quantitatively evaluating the candidate materialized views; deciding whether to approve and deploy the candidate materialized views; and conducting a health assessment of deployed materialized views, performing corresponding optimization operations based on the assessment results. This achieves automated and precise selection of high-value optimization targets, reducing reliance on senior experts. Quantitative evaluation of candidate materialized views ensures that deployed materialized views provide clear benefits. Furthermore, multi-dimensional health assessments of deployed materialized views automatically identify and eliminate inefficient materialized views, enabling continuous and automated optimization and recycling of computing and storage resources.

The application relates to the technical field of data platform scheduling control, cloud native resource arrangement and intelligent operation and maintenance, and discloses a high-concurrency heterogeneous computing power preemptive scheduling method and system based on an internalized intelligent agent. The method receives standardized event loads, identifies emergency fault tasks based on a composite emergency judgment rule, executes suspension, freezing or checkpoint saving on low-priority tasks, recycles and remaps heterogeneous resources such as video memory slices, execution queues, cache areas and high-speed interconnection channels, and generates a treatment result in combination with an open diagnostic model; meanwhile, recovery, feedback and preemptive scheduling audit objects are made explicit under the constraints of a tenant security namespace and a resource isolation boundary.

The application relates to the technical field of data transmission, and particularly discloses a computing power optimization method and system for cloud computing data center transmission, which is used for solving the problem that the existing cloud computing data center cannot balance resource high efficiency and service quality guarantee between security isolation, real-time scheduling and network transmission risk, the method comprises the following steps: hierarchically deploying delay-sensitive and high-throughput services, dividing core and general resource pools and isolating the same, using AHP optimization, DBSCAN abnormality monitoring, random forest prediction and risk regression scoring, and scheduling cross-node resources according to minimum increments; and the system comprises a node management module, an offline resource mode pre-training module, an online strategy adaptive module, a monitoring module, a priority evaluation module, a performance prediction module, a risk evaluation module, a scheduling execution module and a path control module; and the application is based on hierarchical deployment and resource isolation, and adjusts resources and networks through abnormality detection, random forest prediction and multivariate regression risk evaluation.

This invention discloses an end-to-end slicing resource isolation and QoS guarantee method for 5G private networks, comprising the following steps: Two UPF network elements are set up in the 5G core network. One UPF network element is deployed as a central network element in the centralized control plane of the 5G core network for global management and policy control, while the other UPF network element is deployed as a downstream network element at the edge of the 5G core network for local data offloading. The central network element communicates data through a trunk port, and the downstream network element communicates data through a private access port. Several dynamic end-to-end slices are established in the 5G core network and on the user side, with each end-to-end slice allocated independent trunk ports and private access ports. TE-LSPs are deployed between adjacent routing nodes in the end-to-end slices. Data from the 5G core network is forwarded to the corresponding end-to-end slice through the trunk port and forwarded to the local user through the resource isolation channel and the private access port. This invention can overcome the shortcomings of existing technologies and improve the QoS level of network slices.

This invention provides a method for orchestrating and executing AI task flows, comprising: user-developed code for custom operators; encapsulating the custom operators into images; registering the images into an operator library in a task flow orchestration and scheduling platform; user-configured task flow orchestration and execution parameters via a drag-and-drop interface using operators; orchestrating multiple nodes of upstream and downstream tasks and the execution parameters of each task into a task flow, which is then executed in the form of containers; and returning the execution results to the web interface for visualization by the user after the containers have finished running. This invention uses containers to host task execution, allowing for on-site resource allocation during task execution, achieving greater resource utilization and isolation. Furthermore, this invention supports user-defined task operators for task flow orchestration, enabling more flexible application scenarios for enterprises.

The application provides a database instruction execution method and system for a power system, relates to the technical field of power system database management, and comprises the following steps: obtaining a database operation instruction and a target data object; performing semantic analysis on the instruction to construct an access dependency graph; identifying a cascading influence range according to the dependency graph, and determining a data object set that needs to be synchronously locked; decomposing the operation instruction into multiple sub-instructions and assigning execution priorities to the sub-instructions; assigning resource isolation identifiers to the sub-instructions, limiting data access permissions, and executing. The method effectively avoids data operation conflicts by identifying the cascading influence range of electrical correlation, and improves the concurrent performance and safety of power system database operation.

This invention provides a method for secure startup and end-to-end authentication of DCS controllers based on trusted computing, relating to the field of industrial control system security. The method includes performing layer-by-layer integrity measurements on the controller's startup firmware and software components to generate a set of hierarchical measurement values, and periodically performing memory checks during operation to generate runtime measurement values. Deviation measurements are then calculated and encapsulated as bi-state credentials. Simultaneously, relevant measurement values ​​are linked by a timestamp-based chained hash to generate an evidence chain. During inter-controller communication, credentials and evidence chains are exchanged. By calculating negotiated measurements and adaptive thresholds, resource allocation ratios are dynamically determined, and processor time slices and memory space are allocated to and executed for peer control commands. This invention achieves dynamic trusted authentication and resource isolation throughout the entire lifecycle of the controller from startup to operation, improving the overall security of the DCS system.

This disclosure provides a method and system for training federated learning models based on a container orchestration platform, applicable to the field of federated learning. The method includes creating isolated containerized runtime environments for each client using the container orchestration platform; dynamically selecting a subset of clients as participants in each iteration and distributing the global model; each client training its model locally in its isolated environment based on local data, applying differential privacy enhancement processing including gradient pruning and noise addition, and uploading the updated local model; and the server aggregating the updated local models to generate a new global model. The container orchestration platform provides automated resource scheduling and multi-tenant coordination support throughout the process. This invention constructs a dual privacy protection system by combining algorithm-level noise perturbation with system-level resource isolation, while leveraging the elastic management capabilities of the container orchestration platform to achieve efficient, scalable, and easy-to-maintain multi-tenant federated learning.

The application discloses a kind of distributed real-time indexing method and retrieval system for large-scale data, for maintaining the accuracy of cluster node survival detection when hot spot index block continues high frequency update.Method by real-time acquisition node running index, including hot spot index block update frequency, memory heap usage and garbage collection information, and based on update frequency and memory pressure self-adaptive adjustment memory data flush to solid state disk frequency and batch size, to reduce write load.Further, for the process or thread of execution node survival detection process, set high priority scheduling and divide independent CPU quota, to ensure that still obtain minimum time slice under high load.At the same time, for the detection operation configuration independent from runtime environment heap Memory buffer, and in isolation thread is executed, to complete node heartbeat detection within preset time limit and return response based on CPU and memory resource isolation guarantee.

The present application relates to the technical field of database resource management, and particularly relates to a dynamic allocation method of tenant isolated resources, an electronic device and a storage medium, the method comprises the following steps: in response to a database registration request of each tenant, allocating a temporary isolated space and a tenant identifier to a tenant that passes verification, continuously collecting behavior data of the tenant during the use of the database resources by the tenant, calculating the influence weight of each behavior dimension in the behavior data on the consumption of the database resources, combining the normalized values of each behavior dimension of the tenant to calculate a behavior compliance score of the tenant, so as to dynamically determine the current security level of the tenant, and when a trigger condition of isolated resource allocation is monitored, taking the temporary isolated space of the tenant as a benchmark, dynamically adjusting the isolated resource quota of the tenant according to the current security level of the tenant; the present application can guarantee the service quality of a high-trust tenant and limit the resource abuse of a low-trust tenant, so as to realize effective resource isolation and fair scheduling between tenants.

This invention belongs to the field of container management technology and discloses a container management platform, including: a module that enables collaboration among various modules through an event bus and API gateway; an application high-availability module that relies on a two-layer replica mechanism for rapid fault response; a resource integration module that performs cross-environment resource pooling and real-time monitoring and LSTM traffic prediction for elastic scaling modules, dynamically matching resource supply and demand, significantly improving the utilization of resources such as CPU and memory, and avoiding idle waste; a cost accounting module that calculates resource costs accurately in real time and synchronizes them to the scheduling module, facilitating cost optimization; and a permission and GPU management module that adopts an RBAC+ABAC fusion architecture and NVIDIA MIG technology to achieve fine-grained permission control and isolated sharing of GPU resources, improving the efficiency of heterogeneous resource scheduling. This invention reduces coupling through a distributed architecture, shortens problem localization time through a comprehensive monitoring system, and fully supports enterprise digital transformation.

The application discloses a private cloud-based server resource scheduling method and system, and relates to the technical field of server resource scheduling. The method comprises the following steps: performing multi-dimensional resource modeling on physical server nodes in a private cloud cluster, cutting a physical GPU into a plurality of virtual GPU instances through virtualization technology, and constructing a unified resource pool; receiving a computing task submitted by a user, analyzing the QoS level identifier and multi-dimensional resource requirement of the task; based on the real-time state of the unified resource pool and the multi-dimensional resource requirement, adopting parallel optimization to solve a globally optimal task placement scheme for each candidate node, and generating a scheduling decision; according to the scheduling decision, allocating and binding physical CPU cores and virtual GPU instances for the computing task on a target node, and configuring corresponding resource isolation parameters according to the QoS level of the task. The technical problem of low overall utilization of server resources in the prior art is solved, and the technical effect of improving the overall resource utilization of the server is achieved.

The application discloses a lightweight virtualization isolation method and system based on a Rust token mechanism, and the system comprises a token management module and a trap management module, and the method comprises the following steps: obtaining a resource requirement configuration file, using the token management module to cast corresponding tokens to obtain a token structure body, and transferring the ownership of the tokens to a VM object; consuming Cpu tokens and completing the initialization of CPU context, so that a Guest OS starts to execute; when a synchronous exception trap is triggered, the trap management module obtains corresponding exception tokens and matches strategies to process; when a super call HVC trap is triggered and management operation is performed, the trap management module obtains corresponding HVC tokens and management tokens and performs verification, and after the verification is passed, corresponding management instructions are executed. The tokenization resource isolation method can guarantee high security, reduce runtime overhead, and has good verifiability, portability and expansibility.

This application provides a method, apparatus, and device for isolating and exclusiveizing processor cores in a Linux system. The method includes: receiving a resource isolation configuration request based on a control interface in the proc file system, the resource isolation configuration request including the identification information of the target processor core; performing a resource isolation operation on the target processor core according to the resource isolation configuration request, the resource isolation operation being used to adjust the process scheduling permissions and system resource access permissions of the target processor core; and updating the system scheduling domain configuration based on the resource isolation operation, the system scheduling domain configuration being used to determine the scheduling strategy of tasks on the target processor core. By dynamically adjusting the isolation and exclusive management of multi-core CPU resources through the proc file system, it supports dynamic adjustment of isolation strategies, selective isolation of system resources, and flexible configuration of load balancing modes according to business needs at runtime. This effectively addresses the CPU resource and real-time requirements under different business scenarios, improving system performance and stability.