18 results about "Elastic scheduling" patented technology

This invention relates to a method for elastic scheduling of AI computing resources based on business awareness, comprising the following steps: predefining a business criticality level BK, and mapping the business criticality level BK to a corresponding weight coefficient V. bk The system automatically identifies the current iteration cycle stage of a task and predefines the corresponding dynamic weight coefficient C for the iteration cycle. It calculates and generates a dynamically changing priority score PF using a calculation function. It calculates and allocates the elastic quota FHQ for each business project, obtaining a total quota AHQ. When an urgent task is received, resource preemption assessment is performed. This invention employs a business-aware method, system, device, processor, and its computer-readable storage medium for elastic scheduling of AI computing resources. This achieves deep integration of business and technology, improves resource utilization, effectively suppresses users' false reporting of resource demands, thereby enhancing the overall utilization of the cluster; reduces preemption losses, avoids waste of computing resources, and maximizes the effective computing efficiency of the cluster.

This invention provides a method and system for elastic scheduling of cloud-edge resources in intelligent converged terminals for low-latency scenarios, relating to the field of resource scheduling technology. The method includes: acquiring and decomposing the end-to-end latency constraints of a target low-latency service request, and setting N initial candidate execution paths; mapping the N initial candidate execution paths to path state vectors and performing multi-dimensional evaluation analysis; performing path iterative search of the N initial candidate execution paths based on the comprehensive evaluation vector; configuring an elastic scheduling strategy using the path iterative search results; and performing intelligent scheduling management of the target low-latency service request. This invention solves the technical problems of existing cloud-edge-device scheduling methods that rely on static rules or single-node load, leading to unstable latency and low resource scheduling efficiency for low-latency services. It achieves the technical effect of improving the latency stability and resource scheduling efficiency of low-latency service requests through multi-dimensional constraint decomposition and path iterative optimization to realize cloud-edge-device collaborative scheduling.

The application discloses a tunnel flow dynamic scheduling system for small-granularity slice networks and relates to the fields of network slicing and traffic engineering.The application aims to solve the problems of low resource utilization of small-granularity slices and the difficulty of guaranteeing service experience by passive response scheduling in the prior art.The system comprises a small-granularity slice sensing layer, a dynamic strategy engine and a tunnel execution layer.The sensing layer collects queue depth and time delay data of tunnels in real time;the strategy engine predicts congestion based on a prediction model and identifies idle slices to initiate resource lending; and the execution layer dynamically adjusts paths and queue parameters to complete the lending.The application realizes prediction-based active elastic scheduling, significantly improves network resource utilization and guarantees the performance of high-priority services.

The present application belongs to the field of safety protection, and particularly relates to a cross-border traffic pool dynamic elastic scheduling system and method, comprising: collecting real-time quality and traffic parameters of multiple operator links, combining historical data and implicit intention of business appeal text to generate a scheduling detection sequence, generating state data through fusion calculation and sending to a cross-border traffic pool dynamic scheduling platform; using a traffic pool dynamic scheduling model to decouple the time sequence delay coupling relationship between parameters, predicting the traffic consumption fluctuation caused by cross-domain route convergence, extracting a feature vector containing traffic excess risk probability and link quality fluctuation, generating a target scheduling instruction in response to cross-domain linkage scheduling conditions, and adjusting the traffic distribution ratio or switching the path. The present application realizes elastic resource allocation and excess risk early warning, and improves the utilization rate of cross-border traffic resources.

This application discloses a method and apparatus for rapid deployment and elastic scheduling of model inference services based on declarative workload pooling. The method includes: after a user submits a declarative configuration of the inference resource pool, the inference pool controller constructs a preheating resource pool base and establishes lock-free pooling state management rules based on metadata tags; after a user submits the inference service configuration, the inference service controller matches and locks the target workload through a multi-dimensional pooling scheduling framework, completing activation and service deployment; when the service terminates, the two controllers collaboratively execute differentiated resource reclamation. This invention eliminates the need for an external database, reduces the cold start time of the inference service to the second level, adapts to business tidal scenarios to achieve efficient reuse of computing power, and meets the complex scheduling requirements of a pre-filling and decoding separation architecture.

This application discloses a method and apparatus for dynamic core binding and energy efficiency optimization based on scheduling domains. The method includes the following steps: parsing the CPU topology during kernel startup, identifying and registering the Cluster scheduling domain; constructing a multi-level scheduling domain hierarchy including Cluster, MC, and NUMA layers; periodically detecting the utilization rate of the current scheduling domain, and dynamically switching the scheduling domain hierarchy based on the comparison result of the utilization rate and a preset threshold; and activating an oscillation suppression mechanism when the switching frequency of the scheduling domain hierarchy exceeds a preset frequency threshold. This application optimizes task scheduling efficiency and system energy efficiency through a three-layer elastic scheduling domain architecture including Cluster, MC, and NUMA layers and a dynamic oscillation suppression mechanism. It can achieve fine-grained load balancing, reduce memory access latency, avoid scheduling domain hierarchy oscillations caused by short-term fluctuations, and enhance system stability.

The application relates to the technical field of cloud computing and big data processing, and particularly discloses a big data computing resource elastic scheduling and management system for a hybrid cloud environment, which collects events of each resource node and adds a hybrid logical timestamp to the events to form a global event set; the events are sorted and grouped based on the timestamp, and a set of to-be-scheduled transactions is extracted; the transactions are simulated to be executed in sequence, a resource occupation state graph is dynamically constructed, resource conflicts and cyclic waiting paths are detected, and a consistency verification result is generated; according to the verification result, if there is no conflict, all the transactions are converted into parallel scheduling instructions, if there is a conflict, part of the transactions are intelligently removed based on a dynamic relaxation scoring strategy to resolve the conflict, and a final scheduling instruction set is formed; the instruction set is synchronously issued to each resource node in an atomic operation mode for execution, and state events generated by the execution are fed back to the collection end to drive the next round of scheduling, so that the strong consistency and safety of scheduling decisions are ensured.

This invention relates to the field of cloud computing and intelligent resource scheduling technology, specifically providing a method and apparatus for elastic resource scheduling on a cloud platform. A monitoring and acquisition module collects metrics and topology information of hosts and VMs; a feature preprocessing module performs short-term smoothing, anomaly denoising, and holiday and event marking to provide cleaned input for prediction; a hybrid prediction module outputs the resource demand prediction and confidence interval for each VM within a future window T; a decision optimization module generates candidate action sequences, which are then selected and fine-tuned by an RL strategy; a security policy engine performs compliance and risk verification on the actions; the Executor executes the actions step by step on the cloud platform, performing canary deployments, monitoring checks, and rollbacks during execution; the Execution Coordinator and the Feedback Learning module store execution data, supporting offline replay training and online fine-tuning. Compared with existing technologies, this invention maximizes resource utilization and reduces operational costs while ensuring business continuity.

The application provides an elastic scheduling method based on dynamic grading of multi-tenant monitoring resources. First, multi-tenant index data is collected. When an index update event is triggered, a weighted average update interval is calculated according to the multi-tenant index data. The weighted average update interval is matched with preset grading rules to obtain index grading information. The multi-tenant index data is dynamically collected according to the index grading information, and the elastic scheduling of dynamic grading of multi-tenant monitoring resources is completed. The method can sensitively perceive changes in tenant business load, dynamically adjust resource allocation and collection frequency, and realize dual optimization of resource utilization rate and service stability.

The application discloses a multi-cloud computing power elastic scheduling method for new media production tasks, relates to the technical field of computers, and is used for solving the problems of load balancing or resource configuration in the coarse-grained allocation mode of the new media industry; through progressive lightweight image extraction, static code stream / container features and extremely lightweight time sequence content features are combined to form a media complexity image vector, so that the differences of different content complexities with the same resolution can be encoded into scheduling input, the accuracy of time consumption estimation of the new media production task is significantly improved, and the deviation caused by the coarse-grained rule is avoided.

This invention discloses a virtual power plant system scheduling method based on deep reinforcement learning, belonging to the field of power energy optimization management technology. The method includes: collecting multi-source real-time data from the virtual power plant; constructing a causal graph model and generating causal feature vectors and adjacency matrices; inputting the causal feature vectors and adjacency matrices into a reinforcement learning framework, performing enhancement processing through a causal attention mechanism to generate an optimized scheduling strategy; extracting economic and safety objective function values ​​from the optimized scheduling strategy, constructing a multi-objective optimization problem, and using gradient descent to track equilibrium points and obtain a flexible scheduling scheme; inputting the flexible scheduling scheme into a high-fidelity digital twin model, performing behavioral risk detection during simulation operation, and outputting a risk report and correction parameter vectors. This invention achieves a leap from statistical correlation analysis to causal reasoning, solving the problem of blind spots in scheduling decisions under high-dimensional uncertainty.

This invention discloses a resource elastic scheduling method and system for a cloud computing platform, relating to the field of resource scheduling technology. The method includes the following steps: setting up a cluster of multiple virtual machines and establishing a shared space between adjacent virtual machines. This invention expands or shrinks the target virtual machine by using the shared space and demand load, ensuring the stability and continuity of task operation within the cluster and reducing the negative impacts of resource contention and service interruptions caused by changes in virtual machines within the cluster. Through the setting of a dependency symbiosis model, task data does not need to be directly migrated during data migration, reducing transmission interruptions, data loss, and inaccuracies during task data migration. Furthermore, migration rules facilitate load balancing between two virtual machines within the same shared space, ensuring the stability and continuity of task operation.

This invention relates to the fields of network security and distributed computing technology, and discloses a node elastic scheduling method and system for massive IoT terminal data access. The method includes: real-time analysis of the instantaneous arrival rate of the data stream to be accessed to extract nonlinear fluctuation characteristics; calculation of the geometric volume expansion rate of the data stream during physical memory allocation; construction of a two-dimensional memory topology map mapping the resident core protection processes; and projection of a virtual expanded geometry generated based on the geometric volume expansion rate onto the map; and Boolean interference analysis of the dynamic outer boundary and the core protection processes under coordinate system constraints, generating secure scheduling instructions or memory isolation instructions based on whether the physical addressing space overlaps or conflicts. This invention materializes the burst characteristics of network traffic into the geometric expansion behavior of node physical memory allocation, proactively avoiding the structural collapse risk of addressing overlap in the underlying hardware shared cache area, and constructing a physical immune isolation barrier for edge nodes.

This invention provides a lightweight scheduling method, apparatus, device, and medium for edge computing, comprising: deploying lightweight components on edge devices, including an Edgelet agent and a Containerd lightweight runtime; deploying a global scheduler in the cloud, dividing the edge local resource pool into edge autonomous systems based on the global scheduler, establishing a bidirectional communication link between the global scheduler and the edge autonomous systems, and realizing a cloud-edge collaborative channel based on the communication connection between the global scheduler and the edge autonomous systems; performing hierarchical elastic scheduling based on the edge local resource pool and the cloud-edge collaborative channel, including: edge nodes performing edge-level localized scheduling within the edge local resource pool according to their own load status; and the cloud-based global scheduler performing cross-domain overall scheduling from a global perspective based on the metadata reported by each edge autonomous system, realizing lightweight collaborative scheduling with the cloud managing the global and the edge managing the local.

The application provides a material structure stream generation and adaptive neural network asynchronous non-blocking routing system based on actor elastic scheduling, belongs to the technical field of computer and material engineering integration, and comprises a feedback learning and optimization control unit, a stream generation and intelligent packaging unit, a zero-copy dynamic routing unit, an elastic execution and adaptive arrangement unit and a storage and memory management unit. The application breaks the serial barrier of structure generation and model reasoning by introducing a pipeline parallel mechanism, establishes a stream generation full-asynchronous link of generation-distribution-reasoning, realizes linear growth of throughput, constructs an extremely fast IO channel by using a zero-copy transmission technology in view of the fragmentation characteristics of material structure files, and realizes direct transmission in a kernel state in combination with NIO multiplexing.

The application discloses a multi-stage elastic scheduling method of a gas-electricity coupling energy system considering source-load flexibility coordination, relates to the technical field of energy system optimal scheduling, and comprises the following steps: a multi-energy flow physical coupling model containing a gas turbine, a ground source heat pump and natural gas pipe network dynamic pipe storage is established; equivalent thermal parameters of buildings are identified through a machine learning method fusing physical mechanism constraints, indoor temperature elastic boundaries are determined based on a thermal comfort model of user sensory fuzziness, and load end demand response reserve capacity is calculated; a collaborative reserve pool composed of an electric side energy storage, a load end demand response and a natural gas pipe network dynamic pipe storage is constructed based on the multi-energy flow coupling model and the reserve capacity, and multi-stage elastic scheduling is executed accordingly: in a day-ahead stage, source-load-grid collaborative reserve capacity is evaluated and optimal scheduling is performed; in a post-disaster stage, load hierarchical recovery is performed with the aim of prolonging system survival time and guaranteeing important load power supply.

This invention discloses a simulation resource staggered allocation and scheduling system based on time-series optimization, belonging to the field of simulation computing cluster resource scheduling technology. Specifically, it includes a time-series-aware resource module, a scheduling task classification module, a staggered planning and management module, a task scheduling decision module, and a resource elastic scheduling module. This invention achieves sensitivity quantification by constructing a multi-dimensional feature set, classifying tasks as time-sensitive or insensitive. Simultaneously, it implements precise staggered scheduling for time-insensitive tasks, fully utilizing off-peak periods and time-of-use pricing advantages to effectively reduce node load peaks and improve resource utilization. Furthermore, it combines time-series load prediction to construct a multi-objective optimization scheduling model, solving for the optimal task-node allocation scheme and scheduling sequence, while simultaneously performing feasibility verification and dynamic monitoring and adjustment.