77 results about "Resource combination" patented technology

This invention pertains to a system optimized for reutilizing allocated underutilized or unused allocated DRAM, comprising a first host, a second host, and a resource composer, interconnected via CXL. Both hosts run packaged computing environments (PCEs), which may be containers or virtual machines, and are equipped to handle respective processes, P1 and P2. The resource composer is tasked with receiving data related to P1's memory usage from a kernel module on the first host, identifying underutilized DRAM mapped to P1, and subsequently remapping it to P2's address space on the second host. This process involves the use of CXL.mem commands, which are then translated into appropriate CXL.cache or CXL.io commands for DRAM access based on the mapping.

The invention provides an AI model training acceleration method and system based on computing power service scheduling, and relates to the technical field of computing power service.Firstly, historical training process data of an AI model and structural features of a current training task model are collected, a computing power demand prediction model is constructed, the structural features of the current training task model are input to generate a computing power resource demand distribution sequence, and a computing power demand prediction model is constructed; the method comprises the following steps of: training a demand change curve of each stage of calculation core quantity, memory bandwidth and data transmission rate, screening a matched computing power resource combination scheme from a computing power service cluster according to a sequence, scheduling computing nodes to execute a training task, and collecting actual computing power resource consumption data in real time; and finally, comparing actual data with the prediction sequence, generating a deviation value, and dynamically adjusting the number of enabled computational nodes, thereby realizing accurate prediction and dynamic optimization scheduling of the computing power demand, improving the utilization rate of computing power resources, and accelerating AI model training.

The invention relates to the technical field of resource scheduling, in particular to a building construction resource optimization allocation system which comprises a resource demand prediction module, a progress state monitoring module, a dynamic offset evaluation module, a resource allocation module and an early warning prompt module. According to the resource allocation method and device, through precise matching of the operation process types, the operation sections and the predicted duration time in the construction stage, a staged resource combination list covering the whole construction process can be formed, resource allocation has higher perspectiveness and accuracy, and the efficiency of resource allocation is improved through comparison of the process completion progress and the set resource combination. The progress offset and the resource offset are quantified in time, rapid response to construction site changes is achieved, resource allocation has real-time perception, dynamic correction and priority ordering capabilities by means of an adjustable human list, a mechanical distribution diagram and a material inventory table, the management efficiency and the resource coordination capability of the construction site are greatly improved, and the resource allocation efficiency is improved. And the construction risk caused by plan errors or field changes is reduced.

The invention provides a resource dynamic scheduling method and system, and the method comprises the steps: obtaining the real-time output, gradeability, start-stop records, health states and other operation information of various adjustable resources, and constructing an operation information sequence according to a unified time reference; performing quality verification and completion on the operation information to generate an operation information basic database; extracting a multi-dimensional capability index based on the operation information, and constructing a resource elastic portrait through normalization, time correlation modeling and feature fusion; generating a dynamic adjustable capacity index according to the elastic portrait, and reflecting the real-time adjustment capability of each resource along with the change of the running state; determining an output adjustment interval of the resources according to the capacity index, and forming a priority scheduling sequence; according to a power gap obtained by load prediction and wind and light output prediction, generating a resource combination scheme meeting an adjustment demand and issuing a scheduling instruction; and monitoring the actual output of the resources, updating the elastic portrait and the capacity index when the deviation exceeds the limit, and correcting the scheduling scheme in a rolling manner.

The invention discloses a cloud data server mixed training method and system, which are applied to a cloud data server cluster comprising a plurality of computing resources. According to the method, a model structure, a data set, a time delay constraint and an isolation constraint of a training task are analyzed to generate a task portrait, resource configuration, an operation state and an isolation capability of a calculation node are collected to generate a resource portrait, and a co-located interference degree index table of a task and resource combination is constructed based on historical monitoring data. During scheduling, a heterogeneous computing power utilization rate, an estimated training time delay deviation and a co-located interference degree are taken as indexes, a candidate resource allocation scheme is subjected to weighted evaluation to generate a mixed training scheduling strategy, and resource isolation is implemented through container and accelerator multi-instance division. In the operation process, the training time delay and the actual co-located interference degree are continuously monitored, the scheduling weight and the co-located interference degree index are dynamically adjusted according to the deviation, closed-loop optimization of multi-task mixed training is achieved, the heterogeneous resource utilization rate is increased, and time delay default and co-located interference are reduced.

The invention relates to a network resource flexible scheduling method, system and device and a medium. The method is applied to a backbone transmission network and data network intelligent management and control system. The method comprises the following steps: disassembling network resources into a plurality of atomic power units in a virtualization layer, marking dynamic attribute tags for the atomic power units to obtain an atomic power library, and generating a trigger event when a real-time performance index exceeds a preset threshold value; according to the trigger event, the atomic power library and the alarm knowledge graph, carrying out combination optimization on the atomic power units to generate a resource combination scheme, and calling a three-dimensional topology model to carry out pre-verification on the resource combination scheme; and according to the execution scheme, carrying out capacity expansion and shrinkage operation on the atomic power unit through the virtualization layer, adjusting a load balancing weight based on real-time flow distribution, and generating a resource allocation result. According to the method, dynamic cooperative scheduling of network resources is realized, and the service quality and the operation efficiency of a backbone transmission network and a data network are improved.

The invention belongs to the technical field of urban logistics and resource management, and provides a logistics resource scheduling method based on rule guidance and hierarchical greedy, which comprises the following steps: S1, customer demand structured analysis; s2, carrying out multi-dimensional logistics resource attribute modeling; s3, realizing resource filtering through rule guidance; s4, defining a combined path template; s5, obtaining an optimal solution through a hierarchical greedy resource matching algorithm for resource scheduling; comprising the following specific steps: S5.1, carrying out resource level dependence modeling; s5.2, setting a comprehensive matching degree algorithm and a matching priority rule; and S5.3, optimizing a resource combination path based on a hierarchical sequence and a greedy strategy. The method can adapt to different customer requirements to realize the reasonability of resource combination, and can also guarantee better resource matching efficiency.

The present application discloses a virtual power plant resource combination optimization method, which relates to the technical field of virtual power plants. The method includes dividing a line section set into four surfaces facing the ground surface, the left wall surface, the underground surface, and the right wall surface. According to the abnormal ambient temperature values ​​in the four surfaces, the temperature loss energy consumption prediction value of each line section in the line section set affected by the ambient temperature is predicted. The crack length and crack direction of the crack section are statistically analyzed to predict the first energy consumption prediction value of each line section in the line section set affected by the crack length and the second energy consumption prediction value of each line section in the line section set affected by cracks of different inclinations. According to the three energy consumption prediction values, the energy distribution required for the transmission line is reasonably allocated. The virtual power plant resource combination optimization method provided by the present invention can improve the rationality of resource allocation.

The invention discloses a computing power resource determination method, apparatus and device, a medium and a product. The method comprises the steps of obtaining real-time computing power resource data of a natural language demand and a resource library; performing intention analysis on the natural language demand based on the historical performance database, and determining a computing power task instruction; and on the basis of the real-time computing power resource data, performing dual-algorithm collaborative matching on the computing power task instruction, and determining a target resource combination meeting requirements from a resource library. According to the method, intention analysis is carried out on natural language requirements in combination with a historical performance database, fuzziness of intention analysis is eliminated from the source, cooperative matching is carried out through a bi-pass algorithm so as to reduce matching deviation of computing power resources, accurate analysis of the requirements is achieved, and then accurate matching of the computing power resources is achieved.

The invention discloses an education resource sharing method and system based on an education cloud platform, and relates to the technical field of cloud platform education resource sharing, and the system can effectively eliminate semantic redundant resources, reduce the repetition rate of a resource library and improve the retrieval efficiency through a resource collection and purification module; the multi-modal resource combination and adaptation module supports the accurate generation of a resource packet according to the teacher demand and the teaching progress, and enhances the teaching fitness of the resource content; the resource recommendation evaluation module constructs recommendation indexes based on user behavior data, dynamically adjusts resource priorities, and realizes priority sharing of high-quality resources; and the verification and authorization module ensures that the structural integrity and teaching adaptability verification is completed before the resources are shared, disqualified content propagation is avoided, and the teaching experience is improved. The whole system realizes a closed-loop process from resource collection, redundancy removal, combination and adaptation to intelligent recommendation and sharing verification, and effectively supports intelligent sharing and accurate pushing of educational resources.

The invention discloses an enterprise service demand prediction and resource optimization configuration system based on artificial intelligence, and relates to the technical field of artificial intelligence. According to the method, the multi-source heterogeneous data is integrated through the multi-modal data fusion algorithm, the demand change rule is accurately captured through the time sequence prediction algorithm, the demand prediction precision is remarkably improved, and the problem that the demand perception ability is limited due to insufficient data utilization in a traditional method is solved; the resource optimization allocation algorithm is combined with the comprehensive utility function of the resource utilization rate and the customer satisfaction, the resource combination is dynamically adjusted, the resource allocation efficiency is greatly improved, the defect that a traditional resource allocation mode is static and lack of flexibility is overcome, and the user experience is improved by continuously monitoring customer interaction and service feedback data. Model parameters are dynamically updated, a resource allocation algorithm is optimized, continuous optimization of resource allocation is achieved, the problem that an existing method lacks an effective feedback mechanism is solved, long-term efficient operation of a system is ensured, and more competitive operation support is provided for enterprises.

This invention proposes a demand response resource combination optimization method that considers response reliability and main transformer load balancing. By constructing a demand response resource combination optimization model, the reliability of resource response and the load balancing of multiple main transformers are incorporated into the evaluation. A heuristic optimization algorithm is used to solve the problem, resulting in a demand response resource combination. The demand response resource combination optimization model includes: an objective function that minimizes the overall network demand response resource response cost; response reliability constraints of participating resources; load balancing constraints after the demand response resource is connected to the main transformer; and logical constraints representing whether the demand response resource is included. The heuristic optimization algorithm, combined with the objective function that minimizes the demand response cost, sequentially judges the satisfaction of response reliability constraints and main transformer load balancing constraints after a certain demand response resource is included in the response, and adjusts the combination of demand response resources to provide a feasible solution to the problem.

The application discloses a cloud data server hybrid training method and system, which is applied to a cloud data server cluster containing various computing resources. The method analyzes the model structure, dataset, time delay constraint and isolation constraint of the training task to generate a task image, collects the computing node resource configuration, running state and isolation capability to generate a resource image, and constructs a co-location interference degree index table of the task and resource combination based on historical monitoring data. During scheduling, the heterogeneous computing power utilization rate, estimated training time delay deviation and co-location interference degree are taken as indexes to perform weighted evaluation on the candidate resource allocation scheme to generate a hybrid training scheduling strategy, and resource isolation is implemented through container and accelerator multi-instance division. During running, the training time delay and actual co-location interference degree are continuously monitored, the scheduling weight and co-location interference degree index are dynamically adjusted according to the deviation, closed-loop optimization of multi-task hybrid training is realized, the heterogeneous resource utilization rate is improved, and time delay violation and co-location interference are reduced.

The invention discloses a virtual power plant resource combination optimization method, and relates to the technical field of virtual power plants, and the method comprises the steps: dividing a line section set into four facing surfaces of a ground facing surface, a left side wall facing surface, an underground facing surface, and a right side wall facing surface, and according to the abnormal environment temperature values of the four facing surfaces, determining the abnormal environment temperature values of the four facing surfaces; predicting a temperature loss energy consumption prediction value of each line section in the line section set caused by the influence of the environment temperature, and counting the crack length of the crack section and the crack direction of the crack section; and predicting a first energy consumption prediction value generated when each line section in the line section set is influenced by the crack length and a second energy consumption prediction value generated when each line section in the line section set is influenced by cracks with different gradients, and reasonably distributing the energy distribution quantity required by the transmission line according to the three energy consumption prediction values. According to the virtual power plant resource combination optimization method provided by the invention, the resource allocation reasonability can be improved.

The invention discloses a service switching method and device, electronic equipment and a computer program product. The method relates to the field of resource scheduling, and comprises the following steps: acquiring M computing power devices, and acquiring services associated with all the computing power devices to obtain N services; calculating power resource combinations required by each calculating power device during business operation are determined, M calculating power resource combinations are obtained, calculating power fitness data of each calculating power device are determined according to each calculating power resource combination, each calculating power resource combination comprises a plurality of calculating power resources, and M is a positive integer greater than or equal to 1, and M is a positive integer greater than or equal to 1. The computing power resources at least comprise one of the following: memory utilization rate, central processor utilization rate, graphics processor utilization rate and network utilization rate; and determining an operation sequence of the M computing power devices according to the M computing power fitness data, and operating N businesses associated with the M computing power devices based on the operation sequence. Through the method and the device, the technical problem that the resource utilization rate is low when the computing power equipment runs services by utilizing various types of resources in the related technology is solved.

The invention provides a GPU computing power resource scheduling method, system and device and a medium. According to the method, task information of at least one to-be-scheduled task is obtained, a corresponding task portrait is constructed based on the task information, resource state information of each computing node in a GPU cluster is obtained, a corresponding node state is constructed based on the resource state information, and then the scheduling priority of the to-be-scheduled task is determined according to the task portrait and the node state. And then, according to the task portrait, the node state and the scheduling priority, determining a target GPU resource combination used for executing the to-be-scheduled task, so that the task scheduling efficiency is improved under the conditions of large-scale operation of the GPU cluster, various task types and dynamic change of resource requirements. And selecting a target GPU resource combination which can improve the overall resource utilization rate, reduce inter-task interference and shorten task completion time while ensuring that the high-priority tasks are executed in time.

The invention discloses a multi-resource combination constrained plan-oriented two-point automatic path-finding method and system, belongs to the technical field of path planning, and aims to solve the technical problem that the prior art is disjointed from engineering practical application. Comprising the following steps: acquiring geographic coordinates, construction plan indexes and resource state data as input data; forming a multi-resource combination graph model comprising a node network, edges, a resource composition scene and constraint parameters; calculating the comprehensive weight of each resource combination scene of each edge through a plan-oriented cost function; constructing an improved Dijkstra algorithm introducing constraint screening, screening the edges and the corresponding resource combination scenes in the multi-resource combination graph model through the improved Dijkstra algorithm based on the comprehensive weight of each resource combination scene of each edge, and selecting an initial path scheme which meets a constraint condition and has the minimum comprehensive weight to form an initial path scheme; and the optimal path is solved again by improving the Dijkstra algorithm.

The invention discloses an Internet laundry order resource accounting method and system based on LSTM, and relates to the technical field of intelligent decision making, and the method comprises the steps: constructing an LSTM model, and outputting the real-time fatigue index of each physical resource; receiving and analyzing an internet clothes washing order, and generating a process link; screening out available resources of which the current fatigue index is lower than a safety threshold from all physical resources of the physical resource type to form a candidate resource set; selecting a target physical resource combination for the process link based on the resource combination optimization logic; for a high mechanical load link, selecting a physical resource with the lowest current fatigue index; for a high-heat-load or high-precision requirement link, selecting a physical resource of which the current fatigue index has the largest difference from the average fatigue index of the type of resources; and according to the target physical resource combination and the fatigue index, calculating estimated time consumption and energy consumption of each process link.

A user equipment (UE) is described. The UE may comprise a receiver and a transmitter; wherein the transmitter is configured to perform a Type 2 channel access procedure for sensing a channel in the one or more time resources indicated by the Resource combinations field included in the SCI format 2-C.

This invention relates to the field of teaching resource scheduling technology, and discloses a dynamic scheduling system and method for teaching resources based on multi-agent collaboration. The system includes: a basic modeling module for constructing teaching chain units, teaching chain graphs, resource information tables, and matching rules; a constraint generation module for generating a set of constraint information; a disturbance identification module for identifying affected units, affected ranges, and non-scheduling domain states; a candidate generation module for generating a set of effective composite resource candidates; a collaborative evaluation module for generating evaluation results for candidate combinations; a scheduling execution module for determining the optimal composite resource combination and completing local rescheduling; and a state update module for updating execution state information. This invention realizes the dynamic adjustment of teaching resources and the updating of execution state during the teaching execution process.

The invention relates to the technical field of block chains, and discloses a resource voucher distribution method and device, computer equipment, a readable storage medium and a program product. The method comprises the steps of obtaining user attribute information of an expected user matched with a resource provider in response to a resource voucher distribution instruction; acquiring historical browsing data of the expected user for each resource under the condition that the expected user is determined to meet a resource voucher distribution condition based on the user attribute information; according to the historical browsing data, performing interest detection on a target resource browsed by the expected user; under the condition that the interestingness of the target resource meets a resource voucher combination distribution condition, determining an associated resource of the target resource from the resources; generating a combined resource voucher for the resource combination based on the first voucher information matched with the target resource and the second voucher information matched with the associated resource; and distributing the combined resource credential to the expected user. By adopting the method, the resource voucher distribution security can be ensured.

A satellite resource pooling management system based on satellite-ground network integration comprises a resource access and standardization module, a resource state monitoring and global resource view module, a resource pooling scheduling engine, a satellite-ground network integration control module and a service adaptation and feedback module. The resource access and standardization module realizes multi-resource access and resource standardization processing; the resource state monitoring and global resource view module realizes resource state real-time monitoring and global resource view construction; the resource pooling scheduling engine screens an optimal resource combination through a scheduling strategy and a dynamic adjustment mechanism based on service requirements and global resource states; the satellite-ground network integrated control module realizes user terminal cross-satellite / ground station switching and network link optimization; and the service adaptation and feedback module realizes service classification adaptation and closed-loop feedback. According to the invention, pooling integration and integrated management and control of satellite, ground station and network link resources are realized, and the resource utilization rate, the service continuity and the system adaptability are improved.

The invention discloses a hardware port reflection method and system for a real-time simulation test resource combination system, and the method comprises the steps: carrying out the port conversion of hardware resources in an interface processing and simulation test resource control module of the real-time simulation test resource combination system; when the real-time simulation platform receives data or detection signals, bus data or signals are obtained from the electrical interface through the physical layer, then the received data / signals are analyzed through the application protocol layer, variable data needed by simulation testing are obtained, and the variable data are stored in the DMA; the real-time simulation platform directly obtains variable data in a form of accessing a memory address; when the real-time simulation platform sends data or outputs signals, the simulation test execution program directly writes variable data generated by output into the DMA and then sends or outputs the variable data. The system is used for executing the method. The method has the advantages of being simple in principle, wide in application range, capable of improving the real-time performance of simulation work and the like.

Disclosed methods and systems for implementing serverless management of platform resources maintain a function-as-a-service (FaaS) mapping database associating a resource and management function combination with a locator identifying a container image for implementing management interactions required to perform the management function. When a FaaS request indicating a function-resource combination is detected, the applicable locator is retrieved from the mapping database and the corresponding container image is deployed. Each resource may correspond to a specific version of a specific make or model, e.g., a server model and a firmware version. API, protocol, and messaging specifics for performing the function on the resource are embedded in each container image. Thus, the container image for an inventory operation on a server with a first version of firmware may execute SOAP-based messaging while the image for the same operation on the same server model provisioned with a second firmware version may invoke a RESTful API.

A system and method for a using a genetic algorithm with an extra bit in the genome to obtain a best-fit of resource combinations for a particular task to be scheduled. The system can generate a genome from a plurality of jobs and then add an extra bit that can be used to find the best resource combination in the event that multiple resource combinations are equally suitable as an initial fit.

To support "cooperation" in which users provide their resources to each other.SOLUTION: A support apparatus includes: a support resource registration unit which receives resource data input from a first user, the data including a provision available period of a resource, and registers the resource data on a database; a support request receiving unit which receives, from a second user, a support request including a support request period; and a support service determination unit which determines combinations of resources according to the request by referring to the database, and outputs the determined combinations of the resources, as a support service to be provided to the second user.SELECTED DRAWING: Figure 1

This disclosure provides a resource scheduling method and apparatus, electronic equipment, and computer programs. The resource scheduling method includes the steps of: obtaining a plurality of candidate resource combinations corresponding to a target task in response to a resource scheduling request for a target task, each candidate resource combination including at least one type of target resource and the number of target resources required for the target task; for each candidate resource combination, searching a multilevel resource pool in order from low level to high level based on at least one type of target resource and the number of target resources to determine a seed resource pool corresponding to the candidate resource combination; and determining a target resource pool from a plurality of seed resource pools to satisfy the resource scheduling request based on the level of the seed resource pool corresponding to each candidate resource combination. This disclosure is used in the technical field of networking. Embodiments of this disclosure can improve the efficiency and flexibility of resource scheduling.

The invention provides an equipment first-aid repair task planning method, and belongs to the technical field of equipment first-aid repair, and the method comprises the steps: collecting basic parameters, faults, positions and task association information of to-be-repaired equipment, carrying out the standardization processing, and constructing an equipment first-aid repair information set; performing quantitative scoring on each task in the equipment first-aid repair information set, and generating a first-aid repair task priority ranking table; establishing a first-aid repair resource dynamic database, updating resource state information in real time, and matching an optimal first-aid repair resource combination for each first-aid repair task by adopting a priority priority and resource adaptation principle according to the priority ranking table; and in combination with time constraint, space constraint and resource constraint conditions of each first-aid repair task, performing overall scheduling on the matched tasks and resources to generate an initial first-aid repair task planning scheme. According to the equipment first-aid repair task planning method provided by the invention, the first-aid repair response time is shortened, the resource utilization efficiency is also remarkably improved, and the requirements of rapid first-aid repair of equipment in emergency rescue and other scenes are effectively met.

The invention provides a distributed system resource allocation method which can be applied to the technical field of artificial intelligence. The method comprises the steps of performing structured processing on unstructured data in resource related data based on a preset optical character recognition algorithm to obtain an entity association relationship of resources, and performing special analysis on the entity association relationship through preset functional differentiation analysis agents to obtain special analysis data; carrying out dynamic weighted summation on the special analysis data by utilizing a decision-making agent to obtain fusion analysis data, and generating an execution instruction corresponding to the fusion analysis data in response to that the fusion analysis data passes compliance verification; and calling a regulation and modification program interface according to the execution instruction, and carrying out configuration regulation and modification on the distributed subsystem based on the regulation and modification program interface according to resource category allocation in the resource combination optimization scheme. The invention further provides a distributed system resource configuration device and equipment, a medium and a product.

The invention discloses an intention-based heterogeneous computing power resource declaration and automatic configuration system and method, and belongs to the technical field of computing power distribution. The intention-based heterogeneous computing power resource declaration and automatic configuration system aims at solving the problems that in the prior art, heterogeneous computing power resource configuration is complex, depends on artificial experience, is tightly coupled with applications and resources, and is difficult to automatically optimize configuration directly according to an application performance target. According to the method, the system can automatically deduce and configure the appropriate heterogeneous computing power resource combination only by declaring the performance intention of an application developer, so that the configuration process is simplified, the resource utilization rate is increased, and the application performance is guaranteed.