38results about How to "Timely execution" patented technology

The invention provides a system and a method for distributed asynchronous task queue execution in a cloud environment. The system comprises an interface device, a task queue device and a plurality of task execution devices. The interface device is used for receiving a task execution request submitted by a user, verifying whether the quota of the user is overrun, and receiving tasks submitted by the user and writing the basic information of the tasks into a database under the condition that the quota of the user is not overrun. The task queue device is used for judging whether the tasks submitted by the user are tasks needing to be executed in a parallel mode or tasks needing to be executed in a serial mode, and correspondingly putting the tasks submitted by the user in a parallel task queue and a serial task queue. The task execution devices are used for executing parallel tasks and serial tasks pushed by the task queue device, the task execution devices judge whether the degree of concurrency exceeds the limit of the user when the task queue device pushes parallel tasks, and the task execution devices execute the parallel tasks when the degree of concurrency does not exceed the limit of the user. The system of the invention has the advantages of high task execution efficiency and good performance.

The invention discloses a business request adjusting method and device. In a business request processing system, a first queue and a second queue are pre-created. The number of threads opened by the first queue is higher than the number of threads opened by the second queue; and the priority of the business request in the first queue is higher than the priority of the business request in the second queue; when the adjustment time arrives, it is confirmed whether the number of business requests currently being threaded in the first queue is less than a processing threshold, and if so, the business requests in the second queue are transferred to the first queue according to a preset adjustment strategy so as to dynamically adjust the resources allocated to the service requests in the queues according to the usage of the resources, to ensure that the business requests with higher priority are executed in time, and to ensure that the business requests with lower priority are processed as quickly as possible when the resources are idle enough. Therefore, the resource utilization efficiency is increased.

The method comprises: the first bearer controller sends a detection message to the second bearer controller in order to detect the state of bearer network between two bearer controllers; the second bearer controller reports the received detection message to the service controller; according to the detection message, the service controller makes session control. By invention, the service controller can get the state of bearer network in each time interval, and according to the information about the state of bearer network, takes relevant operations.

The invention provides a task allocation system for a parallel ordering cloud service platform, comprising a computing cluster and a cloud platform control module. The cloud platform control module is used for acquiring the hardware parameters of virtual machines on each physical node, the idle rate and the weight score of each physical node, segmenting tasks according to the hardware parameters of the virtual machines on each physical node, the idle rate, the weight score of each physical node and preset parameters, and allocating the segmented tasks to the corresponding virtual machines. The invention further provides a task allocation method for a parallel ordering cloud service platform. The allocation weight of a task is computed according to the load of each part of a virtual machine, the hardware performance of each part of a virtual machine and the hardware performance of a corresponding physical node, and the task is allocated according to the allocation weight. Thus, task allocation is scientific and reasonable, and tasks issued by users can be performed in time.

The invention provides a multi-network message pushing method based on an Andriod mobile phone system. A mobile phone and a server are required to be provided; the mobile phone comprises a message backstage service; the message backstage service is automatically started when the mobile phone is started up; the method comprises following steps that: the message backstage service of the mobile phone of a sender selects the sending mode of an instruction message through judging whether the message backstage service of the mobile phone of a receiver is connected to the server, if the mobile phone of the receiver is connected to the internet, the instruction message is packaged into a network message, the network message is forwarded to the mobile phone of the receiver through the server, otherwise the message is packaged into an encrypted short message, the encrypted short message is sent to the mobile phone number of the receiver through a GSM network; the message backstage service of the mobile phone of the receiver intercepts the instruction message, analyzes the instruction message and executes the instruction. According to the method provided in the invention, different message sending modes are selected according to different network connection states, therefore, the message is ensured to be pushed and executed immediately, the message transmission stability is ensured, and the message processing efficiency is improved.

The invention discloses a classification scheduling method and device adapting to a time-triggered network, and relates to the field of time-triggered Ethernet. The device is additionally provided with a link delay measurement module and a flow control module. According to the method, in the receiving direction, an Ethernet frame classification module divides received Ethernet frames into a TT frame, an LT0 frame, an LT1 frame, a PCF frame, a PT frame and an ET frame and then sends the frames to corresponding modules respectively; in a transmission direction, the modules of the various framessend the corresponding types of frames or indication signals to the Ethernet frame scheduling module; the Ethernet frame scheduling module schedules and sends various frames in a unified manner, and on the basis of ensuring the determinacy and real-time performance of the original time-triggered TT service, automatic and dynamic measurement of link delay is realized from a hardware bottom layer through an LT0 frame and an LT1 frame, so that the measurement precision is improved, and the clock synchronization is ensured; flow control can effectively prevent ET frames from being lost due to insufficient bottom-layer storage space of a receiving end, and the efficiency of network data transmission is improved.

The embodiment of the invention provides a method for accessing a solid state disk and storage equipment, the storage equipment comprises a network card, a memory and the solid state disk, in the method, the network card receives to-be-written data and a write command, the write command is used for instructing to write the to-be-written data into the solid state disk, and after the to-be-written data and the write command are received, the network card writes the to-be-written data into a memory and writes the writing command into a submission queue of the solid state disk, and then the network card informs the solid state disk of the to-be-executed command in the submission queue of the solid state disk and writes the command when the to-be-executed command is executed. And the solid state disk acquires the to-be-written data from the memory according to the writing command, and writes the to-be-written data into a medium of the solid state disk. According to the method, the time delay of data access can be greatly reduced, and meanwhile, the CPU occupancy rate of the storage equipment can be greatly reduced.

The invention discloses an instruction scheduling method and system for a multi-period instruction and a medium. The method comprises the steps that the multi-period instruction and a dependent instruction thereof are recognized after instruction decoding; reading a priority scheduling instruction buffer, if a corresponding record exists in the priority scheduling instruction buffer, judging thata data dependence relationship exists between the current instruction and a certain multi-cycle instruction, and marking the current instruction as a priority scheduling instruction; when the instruction is renamed, writing address information of the instruction on which the multi-cycle instruction depends into a priority scheduling buffer; and preferentially transmitting an instruction marked asa preferentially transmitted instruction when the instruction is transmitted. According to the method, the instruction parallelism characteristic in the high-performance out-of-order microprocessor isfully explored, the multi-cycle instruction is executed as early as possible through instruction scheduling for the multi-cycle instruction, the performance of the processor is improved, the implementation cost is low, and the method can be conveniently integrated into an existing processor design.

The invention discloses a method and a television for providing reminder messages, which solve the problem that in the prior art, the reminding effect of the events in daily life to people is poor. The method for providing the reminder messages comprises the following steps that: the reminder parameters are received, calculation is performed according to the reminder parameters and the preset calculation rule; and the television generates a reminder message and outputs the reminder message according to the calculated result. According to the technical solution, the event which is required to be reminded can be previously determined by the user, and relevant information of the event is supplied to the television, so the reminder of the television can be obtained. Since the application frequency of the television which is used as a conventional household appliance is high, the reliability is high by adopting the television to remind the events of the daily life, thereby favoring in time execution of the pre-determined event of people.

The invention provides a GPU multi-core processor shared computing resource preemption method and device, user equipment and a storage medium. The method comprises the steps that a management scheduling thread receives priorities of threads in a thread pool corresponding to service cores or threads from at least one service core or thread; the management scheduling thread judges whether the priorities of all the threads are the same or not; if so, a step of exiting the preemption processing flow is performed, otherwise, the step 3 is executed; the management scheduling thread acquires the lowest priority in the priorities corresponding to all the currently executed threads; the management scheduling thread judges whether the thread with the priority higher than the lowest priority does notacquire the computing resources or not; if not, the preemption processing flow quits; if yes, the step 5 is entered; according to the sequence of the priorities from low to high, the task of at leastone service core or thread corresponding to the lowest priority which is being executed is selected, and a suspension operation is initialized on the selected task.

The invention discloses a meter reading task execution method based on meter reading task instances. The invention belongs to the technical field of power. The method comprises the following steps: S1, generating a meter reading task instance: when a meter reading task instance generation module is started, reading and analyzing a meter reading task configuration file; generating a meter reading task in a memory, after the meter reading time reaches, combining the meter reading task with each meter to generate a plurality of meter reading task instances, wherein the meter reading task instances comprise information such as instance numbers, meter numbers, meter reading data items, instance starting time, instance expiration time, meter reading task instance states and priorities; S2, storage of meter reading task instances: enabling the meter reading task instance generation module to store the generated meter reading task instances into a meter reading task instance database for persistent storage, so that power failure or process restart loss is avoided, and S3, notification after meter reading task instance updating: after the meter reading task instance generation module generates new meter reading task instances, notifying the meter reading task instance execution module.

The invention discloses an identity data synchronization system and a method based on a timing task, and aims to provide the identity data synchronization system and the method based on the timing task, which are efficient in data synchronization. The system comprises a timing task management module, an application system management module, a synchronization demand management module, a synchronization data processing module, a failure request management module, a synchronization strategy configuration module and a synchronization log auditing module. The timed task management is used for providing a main timed task and an auxiliary timed task; the application system management is used for managing a third-party application system integrated to the unified identity authentication platform,providing an application information configuration function and maintaining information for identity authentication and configuration of identity information synchronization; the synchronization demand management is used for monitoring the change operation of the user identity information in the system in real time, and generating a synchronization demand in combination with the application information opened by the user once the identity information is changed; and the synchronous data processing is used for generating final synchronous interface communication data for initiating an interfacerequest.

The invention discloses a test case scheduling method and device, a computing device and a computer storage medium. The use case identifier of a test case to be scheduled is obtained from at least onetest case identifier queue. A first virtual machine whose current state is an available state is queried according to a task identification of a task to which the test case belongs and a system identification of a test system required for executing the test case. The execution task of the test case is dispatched to the first virtual machine for the first virtual machine to execute the test case.Through the invention, the machine for executing the test case can be directly queried without executing the snapshot operation first for any test case to be scheduled, thereby realizing the fast scheduling of the test case, enabling the test case to be executed in time, and improving the online progress of the software product or the new function thereof.

The invention relates to a data processing method and system. The method comprises that a server receives a first request which comprises first and second flows; a first processor is created accordingto the first flow and used to process the first flow; and after that the first processor processes the first flow, a second processor is created according to the second flow and used to process the second flow. Thus, the user request is divided into the first and second flows, so that the core flow can be processed in priority, and the non-core flow is processed after that the core flow is processed. It is ensured that the core flow can be executed timely. The probability of request processing failure is reduced, and the user experience is improved.

The invention provides an offshore wind power intelligent management method, device and system, and the method comprises the steps: obtaining an emergency degree grade, a needed operation qualification grade and an operation place corresponding to an operation task when the operation task is received; if the emergency degree grade is greater than a preset grade threshold value, taking the marine workplace which is less than a preset distance threshold value from the workplace and is configured with the vehicle as a candidate departure place; judging whether the operation qualification grade ofthe operating personnel in the candidate departure place is higher than the required operation qualification grade or not; determining whether a feasible route exists between the candidate departureplace and the operation place or not according to the marine meteorological prediction information and the stormy wave resistance index corresponding to the vehicle; and if the feasible route exists,performing route and task allocation. On the premise that safety is guaranteed, an operator can successfully arrive at an operation site, an operation task is executed in time, and normal operation ofan offshore wind power plant is guaranteed.

The invention discloses an outbound call method and device, equipment and a storage medium, and belongs to the technical field of voice calls. The method comprises the steps of pulling outbound data from an outbound database through a master outbound service in an outbound service system, forming an outbound task according to the pulled outbound data, storing the generated outbound task in a line slot, wherein the outbound service system comprises the master outbound service and a plurality of slave outbound services; and pulling the stored outbound task from the line slot through each slave outbound service, selecting a target voice central control platform from a plurality of voice central control platforms based on a load balancing strategy, and sending a voice outbound request to the target voice central control platform according to the pulled outbound task, so as to indicate the target voice center control platform to initiate a voice outbound call process. According to the technical scheme provided by the embodiment of the invention, the call-out efficiency can be improved.

The invention provides a charging method and a charging device of an AGV transport vehicle and an AGV system, and the method comprises the steps: determining a target AGV transport vehicle, the target AGV transport vehicle being one of AGV transport vehicles which do not execute a task; the target charging amount is determined at least according to the first power consumption and the second power consumption, the first power consumption is the electric quantity needed by the target AGV transport vehicle from the first preset position to the second preset position, the second power consumption is the average value of the electric quantity consumed by the target AGV transport vehicle for task execution, and the first preset position is the position of the map center; the second preset position is the position of the charging pile; according to the target charging amount, the target AGV is controlled to be charged; and under the condition that the target AGV transporter is charged with the target charging amount, the target AGV transporter is controlled to execute the target task, it is ensured that the target AGV transporter executes the issued task in time, and therefore the problem that in the prior art, the operation efficiency of the AGV transporter is low is solved.