38results about How to "Implement scheduling" patented technology

The invention belongs to the technical field of computing power scheduling, and particularly relates to a computing power scheduling method in a distributed computing environment. According to the task scheduling method and system, the task needing to be processed at present is obtained through the task management system, the task is divided into the real-time task set, the priority task set and the common task set, and the related tasks are scheduled by adopting the corresponding scheduling strategy based on the set corresponding to the current task and the priority in the set. Resource scheduling and allocation in a distributed computing environment are realized; according to the method, through the steps of task classification, real-time resource monitoring, resource demand estimation, priority ranking, reasonable scheduling strategy, load calculation, resource migration and the like, the technical problem of computing resource scheduling and allocation in a distributed environment is effectively solved, timely processing of high priority is ensured, the system load is balanced, and the system performance is improved. And efficient utilization of resources and stable operation of the system are realized, so that the performance and reliability of the whole distributed computing environment are improved.

One or more embodiments of the invention provide a load-aware scheduling method of an inference system and the inference system. The method is applied to the inference system comprising a global scheduler and an inference engine. The inference engine comprises a Prefill engine and a Decode engine; the Prefill engine and the Decode engine respectively comprise at least one calculation instance; computing resources of the computing instances comprise GPUs carried on the computing nodes where the computing instances are located; the method comprises the following steps: a global scheduler obtains a target reasoning request to be executed, determines a first calculation instance in a Prefill engine of which the GPU load meets a first preset condition based on maintained GPU load information, and sends the target reasoning request to the first calculation instance to execute reasoning calculation in a Prefill stage; and a first local scheduler corresponding to the first calculation instance in the Prefill engine determines a second calculation instance in the Decode engine with the GPU load meeting a second preset condition based on the maintained GPU load information, and sends an inference calculation result of the Prefill stage to the second calculation instance to execute inference calculation of the Decode stage.

The invention relates to a bandwidth scheduling method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: determining the excess bandwidth amount of a node to be scheduled according to the cost bandwidth amount and the current bandwidth amount of the node to be scheduled; determining a plurality of associated nodes of the node to be scheduled, and determining respective residual bandwidth amounts of the plurality of associated nodes according to respective cost bandwidth amounts and respective current bandwidth amounts of the plurality of associated nodes; determining affiliation information of the to-be-scheduled node and respective affiliation information of the plurality of associated nodes, and constructing a node relation graph based on each piece of affiliation information; and generating a bandwidth scheduling path for the excess bandwidth amount based on the node relation graph and the residual bandwidth amount of each associated node. By adopting the method, the scheduling of bandwidth resources can be realized.

The invention discloses a paper machine wastewater treatment monitoring system, and particularly relates to a wastewater treatment monitoring technology, the system comprises a pretreatment module, a process section wastewater monitoring module, a wastewater abnormity tracing module, a process section wastewater diversion module, a process section wastewater confluence module, a wastewater treatment pool scheduling module, a wastewater treatment evaluation module and a database. According to the pollution index of the wastewater of each process section monitored each time in history, the pollution index of the wastewater of each process section monitored currently is predicted, and the initial flow guide direction of the wastewater of each process section is set. According to the predicted water inlet amount, the discharge amount and the residual capacity of the wastewater pool of each treatment grade, intelligent automatic scheduling between the wastewater pool of each treatment grade and the buffer is realized, scientific treatment of wastewater to flow into the wastewater pool of each treatment grade is effectively realized, efficient treatment of papermaking machine wastewater is realized, and the production efficiency is improved.

The invention discloses a cloud side end scheduling collaboration method and system for a cloud game. The method comprises the steps of cloud game service application, service receiving and task analysis, computing network service arrangement (computing network resource perception and network quality perception data intelligence injection), computing power resource calculation and network path calculation (cloud edge collaborative network path and game user access network channel path), computing network resource planning (computing resource allocation service calling), and computing network resource planning (cloud edge collaborative network path and game user access network channel path calling). And network computing service scheduling and network computing service plug-in execution configuration issuing are carried out to complete network computing resources, and finally, the cloud game service carries out business deployment online application. According to the method and the system, intelligent arrangement and scheduling are realized, a'computing network unified brain 'is constructed, and global intelligent arrangement, scheduling and optimization of cloud, edge, end and network resources are realized.

The invention relates to the technical field of automatic production, and discloses an HTRB aging test scheduling method and device, electronic equipment and a storage medium. In the method, an upper computer inquires a test station corresponding to a product; the upper computer sends storage location information of the test station to the PLC, so that the PLC controls the mechanical device to place products; the upper computer receives the information that the product is placed in the storage location; the upper computer sends a product code of the product in the storage location to the aging cabinet for the aging cabinet to age the test product; the upper computer sends the information that the aging test of the product in the aging cabinet is completed to the PLC, so that the PLC controls the mechanical device to take out the product; and the upper computer judges whether the product is qualified or not according to the test result of the product and is used for controlling the mechanical device to transport the product to an outbound area by the PLC. According to the technical scheme, the method is used for efficient scheduling of the HTRB aging test, manual work order switching is changed into equipment program switching, the time for replacing work orders for production of products is greatly shortened, and the production efficiency is improved.

The invention discloses a satellite communication system network control service access architecture system and method based on a software bus, and the architecture comprises a management application layer, a core service layer, an access adaptation layer, and a data service layer. The management application layer comprises management and control service software, network management and control business service software, data operation service software and synchronization service software; the core service layer controls information protocol format conversion software and manages information communication protocol conversion software, and provides an interactive interface of the data service layer and the core service layer; the access adaptation layer provides registration management and information input interfaces for software of the data service layer, and provides registration management and information output interfaces for application services for software of the management application layer; and the data service layer configures user operation and parameters and provides a human-computer interaction interface. According to the invention, unified planning and scheduling of network resources and management and access control of users are realized, visual presentation is provided for managers, and network control service experience is improved.

The invention discloses a resource scheduling method and device of a kubernetes cluster, a storage medium and electronic equipment. The method comprises the steps that in response to a resource scheduling request, a first persistent volume declaration and a second persistent volume declaration of a predefined type are created, the first persistent volume declaration and the second persistent volume declaration are associated with a storage pool, the first persistent volume declaration is a native persistent volume declaration of a kubernetes cluster, and the second persistent volume declaration is a non-native persistent volume declaration of the kubernetes cluster. Capacity configuration information is set in the first persistent volume declaration and the second persistent volume declaration respectively, and the second persistent volume declaration supports increase and decrease of the capacity configuration information; associating a pre-created pod node with the first persistent volume declaration; and creating a storage volume based on the first persistent volume declaration, and mounting the storage volume to the pod node. A new Kubertes custom resource type is introduced, a second persistent volume declaration of a predefined type is created, the second persistent volume declaration supports increase adjustment and decrease adjustment of capacity configuration information, and capacity expansion and capacity reduction of the created storage volume are further supported.

The invention discloses a pole-mounted terminal storage battery intelligent detection system and method, and relates to the field of power distribution automation, and the pole-mounted terminal storage battery intelligent detection system comprises a plug-and-play connection module, a data acquisition and processing unit, a wireless communication module, a cloud service platform, and an energy management module. Data acquisition and analysis processing are performed on the storage battery through the data acquisition and processing unit, the data are sent to the cloud service platform through the wireless communication module, a remote control instruction from the cloud service platform is received through the wireless communication module, and automatic management of energy is realized through the energy management module. All-weather intelligent detection of the pole-mounted terminal storage battery is achieved, it is guaranteed that the pole-mounted terminal storage battery is in the optimal working state, the workload of operation and maintenance personnel is reduced, and the pole-mounted terminal storage battery detection system has the advantages of being intelligent, efficient, high in adaptability and high in reliability.

The application discloses a kind of cluster resource scheduling methods based on kubernetes, including the following steps: creating resource pool, setting the label and the stain of resource pool, part of cluster nodes in cluster is pulled into resource pool;Create tenant, allocate part of cluster resource to tenant, become tenant resource;Project is created under tenant, and part of tenant resource is allocated to project, become project resource;K8s resource Namespace is created under project, and part of project resource is allocated to Namespace;Workload is created under Namespace;Dynamic admission control in k8s is used, the creation or update of Pod is monitored, and the label selection and stain tolerance of Pod are set according to the Namespace to which Pod belongs;Pod is scheduled to the cluster node with corresponding label and stain set according to label selection and stain tolerance.The application also discloses a kind of cluster resource scheduling system based on kubernetes.The application realizes the overall scheduling and management of cluster resource, and can effectively improve the utilization of cluster resource.

The present application relates to the technical field of magnetic element modeling and control, and particularly relates to a multi-factor magnetic core loss prediction method based on machine learning. The method first acquires the magnetic flux density data of the magnetic element, extracts its distribution characteristics and shape characteristics, and realizes excitation waveform recognition by constructing a classification model; secondly, a temperature index correction term is introduced based on the Steinmetz equation to establish a temperature-sensitive loss model; further, the entropy weight method and statistical analysis are used to evaluate the importance of multiple factors; an integrated learning model of random forest and LightGBM is used to construct a magnetic core loss prediction model; and a single-objective optimization model is established by combining differential evolution and simulated annealing algorithm with the objectives of minimum loss and maximum magnetic energy transmission. The method improves the generalization ability and prediction accuracy of the model, can adapt to the modeling needs of magnetic core loss under multiple material, multiple waveform and multiple working condition conditions, and is suitable for application scenarios such as high-frequency magnetic element design and power electronic system optimization.

The invention provides a road emergency vehicle scheduling method and device, and relates to the technical field of emergency scheduling, and the method comprises the steps: collecting road data, and generating a road network which comprises a plurality of nodes; screening the nodes to delete end nodes in the nodes, and determining the remaining nodes as target nodes; emergency rescue related information is acquired, and the total time cost of each emergency vehicle under various decision variables is calculated; adjusting the decision variable according to a constraint condition to reduce the time cost, and obtaining a target value of the decision variable; and performing task arrangement and scheduling on each emergency vehicle according to the target value of the decision variable. The total time cost of each emergency vehicle in the road network is calculated, and the decision variable is adjusted according to the constraint condition to reduce the time cost, so that task arrangement and scheduling of the emergency vehicles are realized, and the material transportation efficiency of the emergency rescue system is improved.

The present application discloses a data processing method, device, equipment and storage medium. The method includes obtaining a task flow to be executed, the task flow includes task characteristics; according to the task characteristics, determining the scheduling information of at least two operator nodes that execute the logical operation instruction of the task flow and the logical operation instruction, the at least two operator nodes include a first operator node and a second operator node, and the operator types of the at least two operator nodes are different; according to the scheduling information, the first operator node and the second operator node are called by a preset mixed operator node to execute the logical instruction, and the target task result of the task flow is obtained. According to the data processing method provided in the embodiment of the present application, geographic information services can be effectively provided to users in complex service scenarios by hybrid arrangement of nodes from different sources and different operator types.

The invention relates to the technical field of satellite communication, and particularly discloses a beam scheduling method for a multi-satellite communication system, which comprises the following steps of: constructing a system architecture comprising S low-earth orbit satellites and M ground cells, and aiming at maximizing the average system capacity of the ratio of the total data flow to the number of satellites required for scheduling; establishing a complex non-convex optimization problem limited by the passing time of the satellite, the number of beam pairs and business matrix decomposition; an optimization problem is modeled as a Markov decision process, a proximity policy optimization algorithm is adopted as a core solver, and policy learning is performed through an actor-evaluator network structure; meanwhile, a matrix decomposition technology is used as an embedded module, and a high-dimensional service matrix is decomposed into a low-dimensional configuration matrix, so that the dimensionality of an action space and the calculation complexity of an algorithm are remarkably reduced. According to the method, intelligent selection and scheduling of the ground cell in a dynamic network environment are realized, and finally, the average system capacity is remarkably improved while the data transmission continuity is ensured.

The embodiments of the present application provide an I / O offloading method, device, system and storage medium in a cloud environment. In the embodiments of the present application, a queue component is added to the CPU of the host, and it is innovatively proposed to use the queue component in the I / O processing process. Based on this, a virtual I / O device and a device queue corresponding to the virtual I / O device can be provided for the instances on the host through the queue component; within the CPU, the device queue is used to schedule the I / O requests that occur between the instance and the virtual I / O device; and the original I / O offloading card only needs to be responsible for listening to the I / O requests in the device queue and transferring the data related to the I / O requests. In this way, the device queue in the I / O processing process can be moved up to the CPU, so that in the I / O scheduling process, it can联动 with each component in the CPU, obtain the real-time loads of each core in the CPU, and thus more reasonably implement I / O scheduling instead of blindly scheduling, which can effectively improve the I / O processing efficiency. It should be noted that the "联动" in the original text seems to be a misspelling. It might be "联动" which is not a standard English word. I've translated it as "联动" for now. If it has a specific correct English expression, please let me know and I can adjust it.

The application discloses a sintering machine tail state monitoring method based on machine vision. The method comprises the following steps: S1, preprocessing of the input sintering machine tail image; S2, color gamut conversion of the original image to extract the combustion area in the image; S3, image processing after extracting the combustion area and extraction of the machine tail area; S4, feature value calculation of the extracted machine tail area; and S5, sintering state judgment and monitoring according to the feature value. The application constructs a sintering machine tail state monitoring system through machine vision and image processing related technologies, realizes real-time monitoring of the sintering machine, gives a prompt and a warning to the abnormal part, further realizes intelligent production of the sintering process section, improves the production efficiency of the steel enterprise, reduces the labor cost, and further promotes the intelligent manufacturing transformation and upgrading and intelligent development of the steel metallurgy field in China.

The invention belongs to the field of power system optimization control, and particularly relates to a network-group-user optimization scheduling system based on a double-layer game optimization model, and the system builds a double-layer game optimization model of a power distribution network-production and elimination user group and a double-layer game optimization model of production and elimination user group-production and elimination users. The upper layer of the double-layer game optimization model is a master-slave game model of the power distribution network and the production and elimination user group, and the power distribution network makes a power purchase price and upper and lower standby prices by taking the optimal operation efficiency of the power distribution network as a target so as to guide the production and elimination user group to optimize production and elimination user resources; the lower layer of the double-layer game optimization model is a production and disappearance user group and a master-slave game model of the production and disappearance users, and the production and disappearance user group formulates an electricity purchase price and upper and lower standby prices so as to excite each production and disappearance user to adjust a charge and discharge strategy of energy storage of the production and disappearance user and the electric quantity and standby capacity transacted with the production and disappearance user group. By constructing a double-layer game optimization model, the operation efficiency of each main body is maximized, and reasonable allocation and scheduling of flexible standby resources in the power system are realized.

The invention relates to a power transmission line dynamic current-carrying capacity prediction method based on grid weather forecast, and the method comprises the steps: selecting a calculation representative point on a power transmission line at an interval of 1 km, and selecting a weather grid point closest to the calculation representative point for pairing; using the weather forecast model to output weather numerical forecast weather grid point data, and mapping the weather grid point data into corresponding environment forecast values of the representative points of the power transmission line; and calculating the maximum current-carrying capacity of the whole power transmission line in the future 1-24 hours by using a heat balance equation. According to the invention, the environment temperature and wind speed forecast values of the whole power transmission line can be formed without increasing hardware environment monitoring stations; the meteorological grid point forecast value is mapped to a plurality of calculation representative points of the long-distance power transmission line, so that the calculation result can reflect the environmental condition of the whole line more accurately; and the maximum current-carrying capacity of the power transmission line in the specific time in the future is calculated by using the heat balance equation, so that the calculation accuracy and practicability are improved.

The invention discloses a sewage treatment centralized control monitoring system based on Web configuration and an application device thereof, and relates to the field of computer software engineering, the system comprises a configuration module, a display module, a control module and a public module; the configuration module is used for configuring and generating configuration data of each factory business unit; one plant area business unit corresponds to one sewage treatment plant; the display module is used for rendering and displaying the configuration data of each factory business unit; the control module is used for performing operation control on the measurement point location and the production process point location associated with the field production equipment of each factory business unit; and the public module is used for maintaining normal operation of the configuration module, the display module and the control module, so that centralized monitoring, decentralized control, real-time data sharing, unified supervision, decision making and scheduling of subordinate water plants are realized.

The invention discloses a full-automatic wafer size grinding control device and method, and relates to the technical field of wafer grinding measurement, the full-automatic wafer size grinding control device comprises a wafer information acquisition module, a coarse grinding and fine grinding first measurement evaluation module and a coarse grinding and fine grinding process dynamic adjustment module, through cooperation of equipment such as a two-dimensional scanning gun, accurate acquisition of grinding information is achieved, and the grinding precision is improved. Rough grinding first measurement is carried out according to a ten-thousandth-micrometer measurement and system analysis to evaluate the wafer state; the number of grinding turns is adjusted according to multiple times of measurement and intelligent calculation in rough grinding; fine grinding first measurement is combined with multi-data evaluation, and dynamic adjustment is also performed according to measurement and calculation during fine grinding. According to the device and the method, an MES system and intelligent manufacturing equipment are utilized, accurate evaluation and intelligent adjustment are carried out in all working procedures, various defects of traditional manual grinding are effectively overcome, the wafer grinding precision, efficiency and quality are improved, and wafer grinding is promoted to advance towards intelligent production.

The utility model discloses a material conveying device for building decoration, which comprises a material conveying mechanism and a limiting mechanism, the limiting mechanism is fixedly connected to the front side of the top of the material conveying mechanism, the limiting mechanism comprises a limiting box, and the bottom of the inner cavity of the limiting box is fixedly connected with a motor assembly. The output end of the motor assembly is fixedly connected with a gear, toothed plates are arranged on the front face and the back face of the gear, the inner sides of the toothed plates are meshed with the gear, the outer sides of the toothed plates are slidably connected with the inner wall of the limiting box, and the outer sides of the toothed plates are fixedly connected with connecting plates. According to the material conveying device, the phenomenon that materials cannot be limited in the prior art is changed, the special-shaped limiting rods are adopted to drive the limiting vertical plates to move towards the inner side, the limiting vertical plates conduct limiting treatment on the conveying box, the inclination phenomenon cannot occur when the conveying box runs to a bumpy road section, and the phenomenon that the conveyed materials fall off cannot occur.

The application provides an information processing method and device of a computing power node, and belongs to the technical field of network security. The information processing method of the computing power node is applied to an access module of the computing power node, and the method comprises the following steps: scanning computing power resource information and security configuration information of the computing power node; and reporting a first scanning result obtained through the scanning to a computing power security authentication service entity. The technical scheme of the application can avoid that a malicious node reports false service capacity.

The invention relates to a video monitoring scheduling method for a bus monitoring center, and aims to solve the problems that in the prior art, monitoring personnel cannot observe key moments in time, potential safety hazards are difficult to find through passive monitoring, and video data transmission occupies a large amount of bandwidth. A vehicle-mounted station broadcaster collects data of a CAN bus, a GPS and the like of a vehicle, preliminary calculation is carried out, first data containing key event information is generated, then the first data are transmitted to a rear-end server for secondary analysis, after cross validation is carried out in combination with information of shift arrangement, lines and the like, whether the vehicle is at a key moment or not is accurately recognized, and the vehicle-mounted station broadcaster is used for carrying out early warning. And a monitoring large screen is dispatched to display related videos accordingly. Wherein the rear-end server judges the operation state of the vehicle, and adopts a more strict data monitoring mode for the vehicle in abnormal operation. According to the method, different video request strategies are adopted for different event types, so that the bus monitoring efficiency is remarkably improved, the data transmission cost is reduced, and the operation safety is effectively guaranteed.

The present application provides a GPU operation method, device, electronic device and medium for edge nodes, the method comprising: obtaining task processing information of a GPU task, the task processing information comprising a task identifier, the task identifier being used to indicate the task type of the GPU task; creating a Pod corresponding to the task type indicated by the task identifier; searching for a target edge node according to an edge node search strategy corresponding to the task type indicated by the task identifier; scheduling the created Pod to the target edge node so that the target edge node executes the GPU task. Through the present application, corresponding scheduling methods can be adopted for GPU tasks of different task types, effectively improving the GPU utilization efficiency of edge nodes.

The embodiment of the utility model provides a handheld terminal, and the handheld terminal comprises a terminal housing which is provided with an accommodation space; the controller is arranged in the accommodating space; the positioning assembly is arranged in the accommodating space, the positioning assembly is electrically connected with the controller, and the positioning assembly is used for receiving the positioning information; the wireless communication assembly is arranged in the containing space, the wireless communication assembly comprises a wireless communication module and a wireless communication antenna, the wireless communication module comprises a 5G communication module, the controller is electrically connected with the 5G communication module, the 5G communication module is electrically connected with the wireless communication antenna, and the 5G communication module receives or sends 5G signals through the wireless communication antenna. The information transmission speed of the handheld terminal and the external terminal is improved through the 5G communication module of the wireless communication assembly, so that the scheduling speed of operators is improved, the accident handling efficiency of the operators is improved, and major accidents are avoided.

The invention belongs to the technical field of power system dispatching, and particularly discloses a new energy-containing multi-region power system economic dispatching method based on an improved differential evolution algorithm, and the method comprises the steps: building a multi-region power system joint reserve capacity calculation model; the new energy uncertainty penalty cost of a certain region is calculated according to the multi-region power system joint reserve capacity obtained in the region, the comprehensive thermal power generation cost and the new energy uncertainty penalty cost are used as a target function, and constraint conditions for keeping stable operation of the power system are considered. Establishing a new energy-containing multi-region power system economic dispatching model; an improved differential evolution algorithm is adopted to solve the economic dispatching model of the new-energy-containing multi-region power system, and a solving result is obtained; analyzing an optimal strategy of economic dispatching of the new energy-containing multi-region power system according to a solving result; reasonable sharing and scheduling of the reserve capacity among the regions are realized, so that new energy consumption is promoted, and the operation cost is reduced.

The present invention provides a road right management method, device, electronic device and storage medium, which relate to the technical field of automated guided vehicles. The road right management method can generate a first planned path through the starting point of the automated guided vehicle and the end point of the task, send the first planned path to the automated guided vehicle and make the automated guided vehicle travel along the first planned path, so as to realize the scheduling of the automated guided vehicle. The road authority of the automated guided vehicle is determined according to the positioning information of the automated guided vehicle and the preset constraint logic, and the stopping or driving of the automated guided vehicle is controlled based on the road authority, so as to avoid multiple automated guided vehicles competing for the occupied path resources and causing deadlock, thereby improving the operation efficiency of the automated guided vehicle.

The present invention provides a communication task scheduling and execution method based on priority queue, comprising the following steps: the host computer UI process sends a command to the communication task scheduling process; the communication task scheduling process converts the command to binary data and sends it to the hardware platform; the hardware platform feeds back the data to the communication task scheduling process according to the received command; the communication task scheduling process receives the data and sends the data to the UI process for display by the UI. The present invention can meet the needs of programmers to conduct simulation experiments on the control protection program to verify the reliability of the program, meet the needs of on-site personnel to monitor the system, understand the operation of the system, the changes in the values ​​of key parameters in real time, etc., and ensure the normal and reliable execution of the communication task; the present invention can ensure the easy expansion and low coupling characteristics of the communication task. The present invention adopts a communication task scheduling and execution method based on the priority queue to realize the scheduling and expansion management of the communication task.

The present disclosure provides a task information processing method and device, equipment and storage medium, and relates to the technical field of information. The method comprises the following steps: receiving task data to be processed sent by a first business target, wherein the task data to be processed is generated by the first business target according to initial task data and task flow strategy information, and the task data to be processed comprises flow target information; and sending the task data to be processed to a second business target according to the flow target information, wherein the second business target is independent of the first business target. The method realizes the scheduling of tasks between independent systems, and solves the problem of lack of reasonable scheduling of tasks between systems due to independent operation of business systems in related technologies.