141 results about "Single task" patented technology

The invention discloses a multi-type robot collaborative operation method based on dynamic environment perception modeling, which realizes optimization and dynamic adjustment of task allocation and forms a complete solution by combining three-dimensional environment modeling and a multi-robot collaborative communication mechanism through a multi-modal data acquisition and fusion technology. Multi-modal data fusion improves the comprehensiveness and real-time performance of environment perception, dynamic environment modeling ensures the accuracy and updating efficiency of a model, and multi-robot communication and task collaboration optimize the resource allocation and execution effect; according to the method, the problems of information isolation, single task allocation and insufficient collaborative operation flexibility in a traditional robot system are improved, efficient collaboration and resource optimization among multiple robots are realized in a complex environment according to the proposed technical scheme, the sensing ability, the task response speed and the overall operation efficiency of the system are remarkably improved, and the collaborative operation efficiency is improved. The system is widely applicable to complex scenes such as inspection, rescue and industrial production.

The invention discloses a monitoring system and method for working data of an AGV robot, and belongs to the technical field of artificial intelligence. The method comprises the following steps: preprocessing and integrating the driving track, freight weight, running state and task execution data of the AGV robot to form a robot working data set; clustering analysis is carried out based on driving tracks of all robots, and similar tracks are clustered into one class; constructing a dynamic driving track planning model by using an A * algorithm, and generating a plurality of candidate optimal driving tracks for each task combination; constructing a task demand list, and carding a task sequence logic relationship; performing task allocation on the AGV robot by using a greedy algorithm; generating a task allocation list by using a simulated annealing mechanism; a task instruction is issued to a corresponding AGV robot control system, a task execution process is started, and whole-course tracking is started; and after the AGV robot completes a single task and a round of task sequence, a model self-learning updating mechanism is triggered.

The invention belongs to the technical field of computers, and discloses a computing power dynamic segmentation method and system, computer equipment and a storage medium, and the method comprises the steps: collecting system historical load data and a task execution mode, and predicting a future resource use condition by adopting a machine learning algorithm model; monitoring the overall load data of the system; searching available computing resources in the system; if the overall load data of the system exceeds an overall load threshold value, the resource demand of a single task exceeds a resource tension threshold value and / or the future resource use condition exceeds the computing power of the system, triggering a computing power segmentation decision; matching the compatibility between the available computing resources and resource parameters of a to-be-executed task, and obtaining a compatibility result; according to the compatibility result, the available computing resources and resource parameters of the to-be-executed task, a segmentation strategy is formulated; and redistributing the available computing resources according to the segmentation strategy. The technical problems of low resource utilization rate and long waiting time existing in computing power segmentation can be solved.

The embodiment of the invention provides a video analysis method and device, electronic equipment, a storage medium and a program product. The video analysis method comprises the following steps: issuing a video analysis task based on preset task arrangement; selecting a target calculation engine from the calculation engines based on the convergence information of the scheduling module and the task arrangement information corresponding to the video analysis task; and performing feature analysis on the video stream based on the target calculation engine. The scheme of the embodiment can be suitable for different scenes and environments, the maximum utilization of environment resources is realized by determining the target calculation engine, and a series of problems of low video analysis capability, single task scheduling capability, low resource utilization rate and the like in an articulated naturality web analysis system are solved.

The invention provides a safety measure generation method and system based on a power distribution network multi-task parallel work ticket, and the method comprises the steps: receiving a to-be-executed work ticket containing a plurality of parallel tasks, and extracting the task information of each task; obtaining a plurality of task groups and task group associated equipment sets corresponding to the task groups; obtaining a task group minimum operation sequence set; obtaining a task group topology boundary equipment set, and performing topology search by using a topology boundary equipment set subset to obtain a safety measure set of each task group; and based on the minimum range maintenance decision model, optimizing the safety measure set of each task group to obtain a final multi-task parallel safety measure set. According to the scheme, the tasks which are electrically connected adjacently are combined into one task group, so that repeated operation on the same equipment can be avoided, unnecessary workload is reduced, and the safety measure generation efficiency of the safety measure generation method based on the multi-task parallel work ticket of the power distribution network is improved.

The invention discloses a task scheduling method which comprises the following steps: receiving a task, scheduling the task or a task fragment of the task to a first execution node in a distributed cluster, and when the first execution node executes the task or the task fragment, scheduling the task or the task fragment of the task to the distributed cluster; and receiving a fragment adjustment request sent by the first execution node according to the load of the task or the task fragment on the first execution node, and according to the fragment adjustment request, adding the task fragment of the task on a second execution node in the distributed cluster, or reducing the task fragment of the task in the first execution node. According to the method, self-adaptive adjustment of the task or the task fragment in the running state can be realized, and resource use is adjusted by dynamically adjusting the task fragment of the task, so that the overall cluster resource achieves load balance, and the risk of task failure or node downtime caused by the fact that an execution node is completely occupied by a single task or the task fragment is avoided.

The invention discloses a distributed layout information reading method and system, a storage medium and electronic device.The distributed layout information reading method comprises the steps that a task configuration file is read through a master control node, the task configuration file comprises layout information configuration, and the layout information configuration comprises a layout processing area and a single task processing area; the main control node is used for dividing the layout processing area according to the single-task processing area to obtain a plurality of sub-task areas, and each sub-task area corresponds to one sub-task; adopting the master control node to respectively distribute the plurality of subtasks to the corresponding slave nodes according to the distribution strategy; the slave node sends a layout reading request corresponding to the subtask to a target database based on a load balancing mechanism; returning layout information corresponding to the layout reading request to the slave node through the target database; and performing corresponding task processing on the layout information through the slave node, and returning a processing result to the master control node. The layout information reading efficiency can be improved.

The invention discloses a Presto task scheduling method and a Presto task scheduling device. According to the method, after a task request is received, tasks are allocated to matched clusters according to user information and task information in the request, then task types are further analyzed, execution modes of the tasks are determined, whether the tasks belong to task types needing to exclusively share resources or not is judged, and if the tasks belong to resource exclusive tasks, the tasks are allocated to the clusters. Tasks are allocated to reserved single tasks and special resource nodes, and when the tasks exclusively share working node resources, no other types of tasks are executed on the nodes, so that the tasks exclusively share the resources on specified cluster nodes, physical isolation of the working nodes in a cluster is realized, the performance problem caused by resource contention is avoided, and the service life of the cluster is prolonged. Therefore, the method is suitable for a complex query environment shared by multiple users, efficient execution of the important tasks can be ensured even when a large number of tasks are processed, and the overall reliability and stability of the system are improved.

The invention relates to the technical field of cache management, in particular to a cache management system based on data monitoring and analysis. The system comprises a data monitoring module, a cache strategy configuration module, a cache execution module, a primary correction module, a performance monitoring module and a secondary correction module. According to the method, the corresponding initial strategy identifier is determined for each task instance, the initial priority and the initial cache sequence of the single task instance are determined according to the single historical loading duration of the single task instance in combination with the corresponding initial strategy identifier, and the initial cache set is generated; and performing primary correction on each initial cache sequence and the initial cache set according to the real-time behavior data and the real-time change data, and collecting a performance monitoring result in a cache execution process so as to update the initial strategy identifier to realize secondary correction on each initial cache sequence. The cache content can be dynamically adjusted, the waste of cache resources is avoided, and the load of the database is reduced.

The invention discloses a multi-heterogeneous edge node-oriented task allocation method and device, equipment and a medium. The method comprises the following steps of: decomposing a single task into a plurality of sub-tasks; when a plurality of equipment nodes are grouped, constructing a variance of execution time of each group of equipment nodes based on a preset grouping number, and constructing a target function by using the variance and the throughput of all grouped equipment nodes by taking the execution time balance of all the subtasks executed by each grouped equipment node and the maximum throughput of all the grouped equipment nodes as a target; and respectively setting constraint conditions for each device node and each subtask, determining groups of the plurality of device nodes and the corresponding subtasks by utilizing the target function and the constraint conditions, and scheduling the corresponding subtasks for each device node. According to the method and the device, when the sub-tasks are scheduled for each device node, each group of device nodes can simultaneously process all the sub-tasks in parallel, and the execution time is kept balanced.

The invention discloses a data acquisition method and system, and the method comprises the steps: obtaining an acquisition task submitted by a user based on an API interface, and carrying out the task analysis of the acquisition task, so as to split the task into a plurality of acquisition subtasks; determining a corresponding task template based on the type of the acquisition sub-tasks, and distributing the plurality of acquisition sub-tasks to different acquisition message queues based on the determined task template; monitoring each acquisition message queue in real time by utilizing an acquisition program, acquiring an acquisition sub-task in each acquisition message queue by utilizing at least one thread, executing the acquisition sub-task by multiple threads to acquire original data, and writing the acquired original data into a summary queue; and consuming the original data by using a consumption program, analyzing the original data into structured format data according to a preset rule, and storing the structured format data into a database. According to the method, a single task is decomposed into a plurality of subtasks and concurrent execution is realized, so that the data acquisition efficiency is greatly improved.

The invention discloses a multi-task log data analysis method and device, equipment, a storage medium and a product, and relates to the technical field of log analysis. The method comprises the steps that a preset multi-task log analysis data set and a preset dialogue template are fused; performing multi-task hybrid tuning on the large language model based on the processed log data set and a quantized low-rank adaptive technology; and inputting to-be-analyzed multi-task log data into the adjusted and optimized large language model to obtain a log analysis result. According to the method, the log data set containing multiple tasks is subjected to template processing, then multi-task mixed tuning is performed on the large language model according to the processed log data set, and efficient tuning is performed by using a quantized low-rank adaptive technology. By means of the dependency between the tasks, the performance of a single task can be improved, correlation analysis between multiple tasks can be carried out, therefore, complex problems can be accurately solved and processed, and the troubleshooting efficiency is improved.

The invention discloses a communication computing parallel optimization method, a multiprocessor system, a medium and a program product, and the method comprises the steps: configuring a first computing core and a second computing core on a single task flow for a general matrix multiplication subtask allocated to a single processor; wherein the first calculation core is used for executing a general matrix multiplication subtask, and the second calculation core is used for executing a set communication task; the set communication task comprises a full accumulation operator or a protocol dispersion operator; then, starting scheduling is conducted on the first calculation core and the second calculation core according to a preset dependency mechanism, so that the general matrix multiplication subtask and the set communication task are executed asynchronously in an overlapped mode; according to the method, the problem of poor reusability of the original Kernel caused by intrusive modification of the GEMM or re-implementation of the Kernel can be effectively avoided through parallel optimization of GEMM calculation and ensemble communication operation, and the performance overhead of the processor is reduced.

The application relates to a multi-round scientific research workflow evaluation system and method for a multi-modal intelligent agent, and belongs to the technical field of large language models and multi-modal large models. The application comprises a multi-round task generation module, a multi-modal information processing module, and the like. The multi-round task generation module constructs a multi-round task scene closer to a real scientific research process. The multi-modal information processing module enables the measured intelligent agent to complete paper active retrieval, multi-modal evidence extraction alignment, cross-paper evidence integration, and round reuse of evidence. The protection points of the application are not limited to a single task sample or a single scoring index, but are focused on a multi-round evaluation benchmark constructed around a real scientific research workflow, a process-level evaluation protocol matched with the benchmark, and an overall implementation scheme supporting landing execution and fine-grained diagnosis of the benchmark.

The invention discloses a position discrimination method and system for multivariate relation auxiliary phrases, and the method comprises the following steps: pre-training a UIE backbone network model which is used for generating first structure data according to task prompts and corpus information; and inputting the first structure data into a multi-task attention network, establishing an association relationship between single task prompt information and text fragment information, and judging the position of a phrase corresponding to the task prompt information in a corpus by using a positioning network and a judgment network. According to the scheme, the relation between the multi-task prompt and the multi-corpus fragment can be processed by combining the output of the UIE backbone network model, the multi-level label attributes can be better predicted, and meanwhile the calculation cost can be better saved.

The invention relates to the technical field of automation control, and discloses a PLC single task to multi-thread task realization method and system, and the method comprises the steps: obtaining PLC task arrangement information which comprises at least one PLC task and a corresponding program organization unit POU; based on the PLC task arrangement information, creating a real-time operating system (RTOS) thread corresponding to each PLC task in the RTOS, and distributing a corresponding stack space for each RTOS thread; when the RTOS task scheduler executes task switching, calling a preset task switching hook function to carry out stack pressing storage on the additional context of the PLC program unit corresponding to the currently running RTOS thread, and carrying out elastic stack recovery on the additional context of the PLC program unit corresponding to the to-be-running RTOS thread, according to the method for transforming the PLC single task into the RTOS-based multi-thread task, the practical requirement that multi-thread task transformation needs to be carried out due to the fact that the real-time performance of a single task mode of a traditional PLC manufacturer is limited is met.

The present application belongs to the field of equipment operation control, and particularly relates to an equipment operation energy consumption control method and system. For the single task software mode of the equipment, a timer of periodic interruption determined according to the required operation period of the equipment and the required load rate of the equipment is set, and the timer count value is read after each time the function module is operated in series to recalculate the main frequency of the next equipment operation to make the energy consumption relatively low and ensure the task to be completed. For the multi-task software mode of the equipment, the self-balancing task priority is set to the lowest, and the operation time of the self-balancing task is used to judge the equipment in the multi-task mode to recalculate the main frequency of the next equipment operation to make the energy consumption relatively low and ensure the task to be completed. Thus, in the single task software mode or the multi-task software mode of the equipment, the equipment energy consumption can be dynamically adjusted, the corresponding task is completed with a relatively low main frequency on the premise of ensuring the normal operation of the equipment, and a relatively low equipment operation energy consumption is achieved.

The present invention provides a general description language data system for directed acyclic graph automatic task flow, including: Step definition layer, Workflow definition layer and Template definition layer; The Step definition layer is the description of a single task, for the input and output declarations of each docker image or other executor, comprises name, type, file and parameters. The Workflow definition layer is a workflow composed of one or more Steps, the dependency topology of these Steps needs to be defined, and shared parameters can also be defined. The Template definition layer is based on a Workflow definition layer. The Template definition layer pre-sets the parameters, and supplies the descriptions, checkers or data source definitions of the parameters. The data center of the present invention is used with the task execution tool, and a programming language needs to be used to implement the corresponding tool.

Systems and methods described herein provide a computer-implemented method for process simulation using a multi-agent system. A software agent is trained using historical process data associated with a process. Each software agent represents a single task in the process. Based on the candidate process model, the process is simulated by executing the multi-agent system to generate simulation results. The software agent operates autonomously within the multi-agent system based on the trained behavior. The process is also performed using a runtime configuration based on the candidate process model. Execution of the process generates runtime results for comparison with the simulation results. In some examples, the candidate process model is adjusted based on the comparison, and simulation and execution of the process are repeated using the adjusted multi-agent system and the adjusted runtime configuration, respectively. A deployment process model is generated based on the adjusted candidate process model.

The application discloses a kind of based on multi-task learning's light path transmission quality prediction method, network reliability and prediction model accuracy are considered simultaneously, a kind of multi-task loss function is designed, and model is trained using the way of joint training, further reduce the error and design allowance of model, improve the accuracy of model, while guaranteeing network reliability, realize network capacity promotion.Compared with the traditional QoT estimation model based on single task, the scheme proposed in the patent can improve the accuracy while ensuring the reliability of the model, thereby achieving the purpose of improving the allocation efficiency of optical path modulation format and improving the capacity of optical network.

The invention discloses a computing resource allocation method and system for heterogeneous GPU hybrid scheduling, and belongs to the technical field of computer equipment.The method is characterized by comprising the steps that allocation types corresponding to tasks to be allocated are identified, and the allocation types comprise single task allocation, group task allocation and emergency task allocation; if the allocation type is grouped task allocation, acquiring the priority of each task in the to-be-allocated task group; the first task is allocated to a GPU node according to the priority, so that the GPU node generates an affinity label of the first task and the unallocated task, and the affinity label responds to the identity label of the unallocated task; according to the method, the affinity labels are obtained through the response mechanism, the tasks are distributed according to the affinity labels, and compared with the method that the affinity relations of the undistributed tasks are obtained by traversing a query table and are distributed, the method has the advantages that the distribution time is shortened, and the scheduling efficiency is improved.

The present invention discloses a method, device and related multi-server system for coordinating single-task master-slave programs, belonging to automatic program design. The method includes: seizing the execution right on a preset master-slave coordination server; if the execution right is successfully seized, the information of the original master program of the single task is obtained from the master-slave coordination server, and according to the information of the original master program obtained, the original master program is connected and the operation of the original master program is stopped; the slave program of the single task deployed locally registers its own information on the master-slave coordination server, and runs as a new master program to continue executing the single task. The present invention provides a master-slave backup solution for deploying a single task on multiple task servers, and well avoids a series of problems such as data duplication and data conflict caused by the single task program running on multiple servers at the same time.

The application provides a node energy consumption optimization method, device, equipment, medium and product of a supercomputing cluster. The method comprises: obtaining resource occupation data corresponding to a task flow; predicting unit task energy consumption when the task flow is deployed on each computing node in the supercomputing cluster based on hardware configuration information and the resource occupation data of each computing node; determining a target computing node from the plurality of computing nodes according to the unit task energy consumption corresponding to each computing node, and distributing the task flow to the target computing node for execution; and adjusting physical operating parameters of the target computing node according to task load changes during task execution, so as to reduce overall operating energy consumption and achieve node energy consumption optimization. The two-stage collaborative control of the scheduling stage and the execution stage can realize fine energy consumption control for tasks, and reduce single task execution energy consumption and overall operating energy consumption of the supercomputing cluster.

The embodiment of the invention provides display equipment. In the embodiment of the invention, the display equipment receives a selection instruction of a multi-window template under the condition that the currently played media asset is a screen projection media asset picture, and supports the screen projection media asset which is played previously to be continuously played based on the screen projection instance by multiplexing the screen projection instance in a single task mode in a task stack. And supporting screen projection media resources indicated by the screen projection instruction sent by the user later based on the screen projection instance. And the multi-window layout in the multi-window playing mode is adjusted based on the screen projection instance, so that mutual shielding among the multiple windows is avoided. On the basis of reuse of the screen projection instance, the problem of playing interruption of previous screen projection media resources caused by screen projection signal conflicts in a multi-window mode is effectively avoided.

The invention discloses a method and a system for realizing intelligent control of a tool body based on a multi-agent architecture, and belongs to the technical field of artificial intelligence and robots. The method aims to solve the problems of single task processing capability, poor robustness and difficulty in expansion of the existing single-body intelligent model. The core thought of the method is to introduce a multi-agent collaboration architecture, decouple'high-level task planning 'and'bottom-level skill execution', and decompose complex control tasks to specialized agents with different functions. The method comprises the following steps: receiving a high-order task instruction; decomposing the high-order task instruction into a plurality of subtasks by a planning agent; the planning agent assigns at least one corresponding execution agent to the subtask from a set comprising a plurality of specialized execution agents according to the attribute of each subtask; and the assigned execution agent cooperatively executes the allocated sub-task so as to complete the high-order task. According to the invention, the modularization, the expandability and the task execution robustness of the intelligent system are obviously improved, and the method is suitable for various intelligent scenes such as industrial manufacturing and home service.

The invention relates to a space crowdsourcing task allocation method based on trajectory prediction, and the method comprises the steps: constructing an MMLSTM model based on a meta path and a Mogrimer LSTM module in combination with end-to-end training, and enabling the MMLSTM model to be used for predicting a user trajectory; and according to the predicted user track, adopting a dynamic strategy of combining grouping insertion and single task insertion to distribute the published new task to the corresponding user on line. Compared with the prior art, on one hand, by introducing the space-time diagram and the multi-type meta path, the complex space-time interaction relationship between the users can be explicitly and structurally modeled, and by combining with the Mogrifier LSTM, the time sequence evolution can be dynamically captured, so that the prediction of the future trajectory of the user is more accurate; and on the other hand, a dynamic strategy of combining grouping insertion and single task insertion is adopted, geographically adjacent tasks are packaged and allocated, and an access sequence is optimized, so that the total detour overhead when a user executes a plurality of tasks can be remarkably reduced.

The application provides a kind of based on self-supervised contrast learning's aspect-level sentiment analysis model training method, system, storage medium and electronic equipment, it is related to natural language processing technical field.Compared with complex downstream structure, the framework only contains pooling operation and fully connected dimension reduction operation, realizes with less additional parameter fine-tuning BERT model.Different from single task fine-tuning mode, the framework takes self-supervised contrast learning as auxiliary task, aspect-level sentiment analysis task as main task, constructs the way of joint fine-tuning to fine-tune pre-training BERT model.Thereby, with the simple downstream structure, more accurate mining of each aspect sentiment in review text is realized.In addition, by means of the characteristics of maximizing similar entity distance and minimizing non-similar entity distance, the BERT model fine-tuned as downstream task can generate more representative features, and effectively assist the aspect-level sentiment analysis task to judge the sentiment conveyed by aspect word entity.

The application provides a fault-tolerant scheduling method and device for an OODAflow operating system, which comprises the following steps: receiving an intelligent computing task and splitting the intelligent computing task into single tasks, encapsulating the single tasks and an environment health degree monitoring process into self-contained containers, and waiting for a computing device to schedule; deploying the single tasks encapsulated into the self-contained containers to corresponding computing nodes; in the process of executing the tasks on the computing nodes, the environment health degree monitoring process monitors the health condition of a computing device to which the computing node belongs in real time, if the health condition has a fault, the computing node exits the task and saves a checkpoint file; if there is no fault, the execution result of the task is written back to a shared storage area. The fault-tolerant scheduling system provided by the application realizes a fault-tolerant mechanism and efficient fault-tolerant scheduling for the OODAflow operating system task, and guarantees the real-time performance and reliability of the OODAflow operating system in processing the computing task.

The invention discloses a fine-grained and high-quality privacy protection data aggregation method. A trust mechanism, a cloud service platform, a fog node, a task requester and a task executor are included. The trust mechanism is responsible for completing generation, distribution and management of a secret key and generation of a pseudonym of a task executor; the cloud service platform is responsible for receiving tasks issued by a task requester, uniformly collecting data and distributing the tasks and information; the fog node is responsible for encrypting privacy information of task executors, distributing selection tasks of an adaptive task executor group and collecting and uploading sensing data; the task requester is responsible for making a task request to the cloud service platform and receiving data collected by a task; the task executor is responsible for executing the task and submitting sensing data to the fog node; according to the method, high-quality data aggregation can be realized, and meanwhile, the location privacy and the data privacy of the user are protected. And the contribution of each user to a single task is considered, so that fine-grained personalized rewards are realized.

Embodiments relate to compiling neural network operations into tasks that can be executed in a streaming manner by a neural processor. In the streaming operation, tensors are spatially segmented, and tasks associated with two or more layers of the neural network are performed simultaneously in an overlapping manner. To achieve efficient memory usage during streaming operations, a subset of tasks whose completion times are temporally close are assigned to the same portion of memory in a neural processor during a compilation process. After the task assigned to the same portion of the memory is completed, the portion of the memory may be refreshed to make space for a subsequent task. Multiple tasks may also be coalesced into a single task to reduce the number of tasks and perform operations more efficiently at the neural processor.