202 results about "Idle time" patented technology

The invention provides a large model training system based on CPU-GPU double optimizers and SSD unloading, and belongs to the technical field of computers. The system comprises a parameter division manager, a CPU optimizer state exchanger, a CPU optimizer, a GPU optimizer state exchanger and a GPU optimizer, large model parameters are divided into a plurality of subgroups through the parameter division manager, and the subgroups are further divided into a CPU updating part and a GPU updating part; a CPU and GPU double optimizer structure is introduced, a double-SSD architecture is adopted to design an independent data path, GPU-CPU collaborative updating is achieved, optimizer state transition between a GPU or a CPU and an SSD is managed, I / O conflicts are eliminated, the GPU idle time is shortened, the influence of SSD read-write high delay is relieved, load balancing is guaranteed, and the training throughput is improved.

The invention discloses a dynamic scheduling method for a multi-satellite collaborative observation task, which mainly comprises the following steps of: dispersing a scheduling period into time slices according to minimum overlapping observation time, and uniformly managing resources and time windows; a multi-objective optimization algorithm is used for initially scheduling fixed tasks, and the task completion rate and the minimum relay frequency are considered; detecting a satellite idle time slice based on the initial scheduling result and constructing an index; the method comprises the following steps of: preferentially performing plug-in scheduling on a temporary task, starting multi-satellite relay if a single satellite cannot complete the task, triggering collaborative rescheduling if the tasks fail, merging the fixed task and the temporary task for re-optimization, and simultaneously minimizing disturbance to the fixed task. According to the method, the resource utilization rate and the dynamic response capability are improved through the strategy of'insertion priority and rescheduling understanding ', the temporary task can be efficiently processed on the premise of guaranteeing fixed task execution, interference on an original fixed task plan is reduced, and the method is suitable for complex and changeable observation scenes.

A power consumption prediction and management method and system based on a solid state disk, a medium and a product relate to the field of solid state disks. The method comprises the following steps: predicting the workload intensity of a host in a first preset time window based on an I / O (Input / Output) instruction stream to obtain an expected host workload curve, and determining an expected host idle time period; calculating an expected host power consumption value in the expected host idle time period; extracting internal state parameters of the solid state disk according to the historical operation log data; establishing a background task triggering probability model, and calculating the triggering probability of starting the high-power-consumption background task by the solid state disk in a first preset time window; performing power consumption conflict judgment according to the expected host power consumption value and the triggering probability, and generating a comprehensive predicted power consumption value; and based on the comprehensive predicted power consumption value, determining an optimal target power consumption state, and instructing the solid state disk to be switched to the optimal target power consumption state. By implementing the technical scheme provided by the invention, the stability of state switching of the solid state disk is improved.

The invention provides a dual-runtime engine collaboration method and system for an industrial edge computing terminal. Relates to the technical field of industrial edge computing. The method is applied to a system comprising a real-time runtime engine, a non-real-time runtime engine, a shared memory module and a resource monitoring module, and comprises the following steps: executing a real-time task and / or a non-real-time task, and performing data interaction between the real-time runtime engine and the non-real-time runtime engine through the shared memory module, periodically acquiring a resource use state of a hardware platform through a resource monitoring module, and dynamically adjusting available resources allocated to an engine during non-real-time operation according to the resource use state; and when the idle time window exists in the scheduling period of the real-time runtime engine, executing the non-real-time task in the idle time window through the non-real-time runtime engine. According to the invention, collaborative optimization of the system on the aspects of time certainty, flexible resource allocation and low-latency data exchange can be realized.

A system and method for universal static multi-transmit CPU design with a static pipeline for automatically parallelizing code is presented. A multi-core and / or multi-processor integrated circuit (2) has a plurality of processing units (21) and / or processing pipelines (53) that simultaneously process instructions for data by executing parallel processing machine code (32). The execution of the parallelized processing code (32) by the parallel processing multi-core and / or multi-processor integrated circuit (2) comprises the occurrence of a delay time (26), wherein the delay time is given by an idle time between the processing unit (21) returning the data after processing a specific instruction block of the processing code (32) for the data and receiving the data required by the processing unit (21) to execute a consecutive instruction block of the processing code (32). The parallel pipeline (53) comprises means for: (i) forwarding by providing a data forwarding from a MEM stage as an EX / MEM register to an EX stage as an ID / EX-stage register; (ii) exchanging by making the results of the Ex-ME-phase registers accessible by the Ex phase of a parallel pipeline (53) to provide a result exchange between the pipelines; and (iii) implementing branch pipeline refresh by providing control conflicts by refreshing only those pipelines (53) dependent on one pipeline (53) based on branch address computation conditions.

The invention discloses a multi-experiment task production scheduling method, device and equipment and a storage medium, and the method comprises the steps: carrying out the sampling of a preset number of times for each production scheduling scheme based on the prior distribution of the duration of each experiment stage, and determining the starting time of each experiment operation, constructing a global time resource scheduling problem under the constraint conditions that the time window constraint of the experimental operation, the time period of the experimental operation belongs to the idle time period of the operation equipment and the time periods of the experimental operation distributed on the same operation equipment are not overlapped, and calculating the ratio of the infeasible times of the global time resource scheduling problem to the preset times; and if the risk value is lower than the risk threshold value, taking the production scheduling scheme as a target production scheduling scheme, and performing online scheduling of the experimental tasks according to the target production scheduling scheme, so that the method is suitable for production scheduling of high-throughput and diversified experimental tasks, and the risk of equipment resource conflict is reduced.

The invention discloses a data pre-reading method, electronic equipment, a storage medium and a computer program product, and relates to the technical field of communication, the method is applied to an operating system, the operating system comprises a general block layer, an IO scheduling layer and a block device driving layer, and the method comprises the steps that in a first idle period of the general block layer, the IO scheduling layer is switched to the block device driving layer; determining pre-reading window information based on the current reading request through the general block layer, wherein the pre-reading window information is used for indicating a pre-reading request of the to-be-pre-read object; applying for a first physical memory for the to-be-pre-read object based on the pre-read window information through the block device driving layer, wherein the first physical memory is used for making cache preparation for the to-be-pre-read object; wherein the first idle time period is a waiting time period for waiting to receive read request completion information after the general block layer issues the IO request corresponding to the current read request to the IO scheduling layer. According to the method and the device, under the condition of not increasing the overhead burden, the data reading rate of the operating system is improved, and the universality is relatively high.

The application discloses a file cleaning method and device, the method is executed by a file cleaning server, and the method comprises the following steps: setting a shared variable which can be accessed by a first component and a second component, the first component and the second component are asynchronous components; enabling the first component to detect a cleaning state of the current time, and determining the value of the shared variable according to the detected cleaning state of the current time; enabling the second component to read the value of the shared variable, and determining whether to clean the files to be cleaned according to a preset file cleaning strategy according to the read value of the shared variable. By introducing the shared variable of the cleaning state of the current time, the application can dynamically identify the time of the executable cleaning operation, effectively utilize the system idle time for file cleaning, and reduce the influence of cleaning a large number of files on actual business during the business peak period. In addition, the second component of the application detects the cleaning state of the current time through an asynchronous thread, and does not affect the execution efficiency of the main process.

The application discloses an intelligent scheduling control method for an automobile part automatic assembly line, and relates to the technical field of automatic control, which comprises the following steps: obtaining the task instruction response time length value and the node execution action state change value of each execution unit in a current scheduling period, and calculating the response delay value and the state fluctuation value of the corresponding node; generating a response efficiency value according to the response delay value and the state fluctuation value of each node, and generating a feedback control level label in combination with the node task attribute information; judging whether to perform compression processing based on the feedback control level label and the available idle time value of each node in the current scheduling period, and if the compression condition is met, encoding the response delay value and the state fluctuation value into a compressed feedback data frame; and completing the relay transmission of the compressed feedback data frame in a time period that meets the scheduling idle condition, so that the scheduling delay pressure control, task state accurate maintenance and instruction issuing sequence dynamic optimization of the automobile part automatic assembly line are realized.

The invention relates to method (200) and computing device (101) for decomposition of one or more computational problems The method (200) comprises obtaining (201) performance information of the one or more network nodes, the performance information comprising idle time of the plurality of tasks hosted on the one or more network nodes, and resource capability of the one or more network nodes. Upon determining (204) that a threshold condition on the idle time of one or more tasks is fulfilled, determining (203) one or more re-deploying configurations for a subset of the plurality of the tasks, based on the idle time information of the plurality of tasks, and on the resource capability of the one or more network nodes; and re- deploying (205), based on one of the one or more re-deploying configurations, the tasks of the subset associated with the one of the one or more re-deploying configurations.

An idle time prediction method for a system includes obtaining n idle durations corresponding to n time points, determining if the n idle durations are of a normal distribution, generating a probability according to m idle states corresponding to m idle durations of the n idle durations if the n idle durations are not normally distributed, selecting a predicted idle state according to the probability, and controlling the system to enter the predicted idle state, where n and m are integers larger than one, and m≤n.

The invention relates to the technical field of big data, and provides a transaction data processing method and device based on idle time, equipment and a medium, which can start a business thread to scan a market queue and a return queue in real time in a current transaction period so as to respond in time. In the continuous scanning process of preset times, when no new market information slice is scanned in the market information queue and no new return information is scanned in the return queue, calculating the current idle time according to an idle time determination strategy after the market information and the return are ensured to be processed; the operator type with high real-time performance is divided into the pre-calculation operator and the decision operator, the pre-calculation operator is processed in the current idle time, the decision operator is processed in the current transaction time, the idle time can be fully utilized, glitch time delay in the transaction process is reduced, and the performance and time delay penetration stability of a transaction system are improved.

The invention relates to the technical field of industrial control system physical layer state monitoring, and discloses an electromagnetic shielding effectiveness control system of a communication equipment connector, which comprises a protocol time sequence synchronization module, an idle noise acquisition module, a state identification and processing module and a decision and interface module, and after determining that continuous noise drift exists, the state identification and processing module firstly suppresses alarm output, communicates with other similar systems in a neighborhood so as to judge cooperative drift, and only releases suppression and outputs an alarm when determining that non-cooperative drift exists. An originally isolated single-point decision which may generate misjudgment due to global environment fluctuation is converted into a system-level diagnosis conclusion which is based on distributed consensus and can clearly distinguish local failure of the connector from global environment change, and the accuracy and reliability of state evaluation of the industrial control system in a complex electromagnetic environment are improved.

A method of performing logical qubit operations (17, 19, 25, 27, 29, 31) on an ensemble of logical qubits (1a-d), each logical qubit having a plurality of physical qubits (3), the method comprising: performing one or more rounds of single qubit logical operations (17, 19, 27, 29, 31) on each of a set of the logical qubits (1a-d); and performing a multi-qubit logical operation (25) using at least two of the set of logical qubits (1a-d), wherein the logical qubits (1a-d) involved in the multi-qubit logical operation (25) are synchronised prior to the multi-qubit logical operation (25) by one or more of: providing periods of physical idle time (21, 35, 37) between rounds of single qubit logical operations (17, 19, 27, 29, 31) performed on one or more of the logical qubits (1a-d) involved in the multi-qubit logical operation (25); and performing different numbers of rounds of single qubit logical operations (17, 19, 27, 29, 31) on the logical qubits (1a-15 d) involved in the multi-qubit logical operation (25) in a fixed period of time.

The embodiment of the invention discloses a casting cleaning production line intelligent scheduling control method based on multi-agent reinforcement learning, and the method comprises the steps: enabling each processing station and a carrying execution unit to make a decision according to an environment state through introducing a multi-agent reinforcement learning frame based on an established state space for casting cleaning production line scheduling, and achieving the intelligent scheduling control of a casting cleaning production line. Through global value function optimization, beat coordination and resource dynamic allocation between stations are realized, workpiece waiting time and equipment idle time are reduced, fixed beat or manual intervention is eliminated, and the overall utilization rate of a production line is improved. Based on the setting of the reward function, the system can automatically optimize the task allocation logic according to the actual operation state, manual adjustment of beat parameters is not needed, the debugging and maintenance workload of the system is remarkably reduced, and manual dependence and debugging cost are reduced.

Systems and methods for determining and reporting actual utilization of a core of a central processing unit (CPU) of a host. Prior to implementation of aspects of the present disclosure, running a poll querying endpoints of a process for work appears to the host's operating system as busy work (e.g., taking full use of the core for the poll duration). However, only a percentage of the duration of the poll is used to process a task of the process, where the remaining duration of the poll is spent querying the endpoints (idle time) and the core is not performing a task. Accordingly, a core utilization reporting system and method automatically detects the processing time of the tasks of a process, determines actual CPU utilization of the core based on a percentage of the time the core is busy polling (doing effectively no work) versus doing actual work (processing a task).

To provide a production system capable of effectively utilizing an idle time which is generated in an entire production line of a mounting substrate due to a difference in work time of work devices constituting the production line.SOLUTION: A production system comprises: a production line in which a plurality of work devices including a printing machine, a mounting machine, a reflow furnace, and an inspection machine is disposed; and a management apparatus. The management apparatus performs: acquisition processing of acquiring a standard work time of each of the plurality of work devices when producing one mounting substrate; extraction processing of extracting a specific work device of a longest work time that is longest in a standard work time; and work setting processing of setting an operation mode of the work devices other than the specific work device to a power saving mode in which power consumption can be reduced more than a standard mode or a high accuracy mode in which work accuracy can be improved more than the standard mode within such a range that the work time does not exceed the longest work time.SELECTED DRAWING: Figure 4

This invention discloses an automatic computer protection system and method, relating to the field of computer file security protection. The technical solution includes: detecting user idle time; identifying sensitive tabs in the browser when the user is about to leave, and extracting the ORB feature point set of their content area as a content fingerprint; creating a Gaussian blur overlay window when the user has left, tracking the positional changes of the content area through feature point matching and affine transformation, and dynamically adjusting the position and size of the overlay window so that the overlay always follows the movement of the file content; and destroying the overlay window when the user returns, restoring the file to normal display. This invention, through content area tracking based on visual features and progressive blurring, solves the problem of sensitive file content being exposed and background tasks being interrupted when a user is briefly away, achieving fine-grained protection at the file and tab levels, and helping to improve the security of computer files in the office environment.

The invention discloses a universal traffic anti-shake processing method based on distributed cache, and the method specifically comprises the following steps: S1, when an event is accessed, storing or updating an event record in the distributed cache according to an event identifier id; s2, obtaining candidate events from the distributed cache in batches; s3, the candidate events are judged according to the idle time threshold value and the maximum waiting time threshold value; when the event exceeds the idle time threshold value or the maximum waiting time threshold value, adding the event into a to-be-processed event list; s4, before business execution, intercepting a business method by a section and injecting the to-be-processed event list into business processing logic; and S5, after the candidate events obtained in batches are processed, performing security cleaning on the event records which are not updated according to comparison of the latest updating time before and after processing. According to the method, the database writing quantity is greatly reduced, and automatic batch merging processing of high-frequency behaviors is realized.

The application belongs to the technical field of computers, and discloses a pipeline checkpoint operation method and an operation system thereof, which comprises the following steps: all computing nodes are synchronously started for checkpoint operation, and the checkpoint operation of the corresponding computing node is performed in the training idle time period of the computing node; wherein the training idle time period of the computing node is the time period when the computing node waits for the feedback of training data from other computing nodes; the checkpoint operation comprises three tasks of serialization, encoding and distribution, the serialization is to gradually convert the current state data of the corresponding computing node into a storable byte stream, the encoding is to encode the serialized data, and the distribution is to distribute the encoded data to a storage node for storage, each task of the checkpoint operation is responsible by different computing resources, and different tasks are executed in parallel in a pipeline mode. Through the above mode, the checkpoint can be effectively saved without affecting the training efficiency.

An electronic device may create a maximum number of frames in a frame buffer queue for a foreground application based on a plurality of frame metrics, detect a scroll operation performed on the foreground application, determine a frame rendering duration of at least one frame based on an idle time on a main thread of the foreground application after detecting the scroll operation, generate an artificial synchronization pulse for creation of additional frames for the foreground application based on the frame rendering duration of the at least one frame, adjust a timeline of a scroll animation created independently of a display synchronization pulse, and display the additional frames by filling the additional frames into the frame buffer queue until the frame buffer queue reaches a maximum capacity.

To provide a system. 【Solution means】 Means for receiving participant information, Means for obtaining schedule information based on the participant information, Means for generating candidate times when all participants are available based on the schedule information, Means for checking the availability of facilities corresponding to the candidate times, Means for presenting the candidate times and the availability of the facilities, Means for analyzing the idle time of public service institutions to efficiently set the schedule of events in the local community, Means for setting a meeting based on the time selected from the presented candidates, Means for sending a notice regarding the set meeting, A system including the above.

The invention discloses a firmware update processing method and device based on AI drive, a terminal and a medium, and belongs to the technical field of intelligent equipment firmware updation.The method comprises the steps that log data and sensor data of an intelligent terminal are obtained, and user use habit data are analyzed through the log data and the sensor data; according to the analyzed use habit data of the user, predicting a future idle time period of the corresponding intelligent terminal by adopting an AI prediction model of a time sequence; according to the predicted future idle time period of the corresponding intelligent terminal, selecting an updating opportunity for confirming firmware updating; and based on the update opportunity of the selected and confirmed firmware update, the arrival of the corresponding opportunity is detected, and the firmware update is controlled to be automatically executed. The method has the advantages of avoiding triggering update when the user uses the equipment at high frequency, improving the update success rate and reducing the system resource consumption, and provides convenience for the user to use.

The invention discloses an intelligent dispatching method for flight departure / departure moments. The method comprises the following steps: establishing a moment distribution calculation model; establishing a flight duration calculation model; judging whether the takeoff moment and the landing moment reported by the airline company meet the flight duration requirement specified by the civil aviation bureau; whether the take-off moment or the landing moment reported by the airline company has idle moments at the take-off airport or the landing airport or not is judged; calculating a suggested moment when the landing moment declared by the airline company cannot be allocated; calculating a suggested moment when the takeoff moment declared by the airline company cannot be allocated; calculating suggested moments when the takeoff moment and the landing moment declared by the airline company cannot be allocated; an airline company calculation result is fed back; and airline company result confirmation. According to the method, the availability can be analyzed by using the preset model when the airline declares the start / fall moment of the flight; and if no time slot exists at the original moment, the adjacent moments conforming to airport flow tolerance and civil aviation duration are automatically recommended according to an airline acceptance range, so that the airline can conveniently reprogram a flight plan.

A system and method for designing a general-purpose static multi-issue CPU using static pipelining of automatically parallelized code is proposed. A multicore and / or multiprocessor integrated circuit has multiple processing units and / or processing pipelines that simultaneously process instructions for data by executing parallelized processing machine code. The execution of parallelized processing code by a multicore and / or multiprocessor integrated circuit that performs parallel processing involves the occurrence of latency, which is given by the idle time of the processing unit between the time the processing unit processes a particular block of instructions in the processing code for data and sends the data back, and the time the processing unit receives the data necessary to execute a successive block of instructions in the processing code. The parallel pipeline includes means for (i) forwarding by providing data forwarding from the MEM stage as an EX / MEM register to the EX stage as an ID / EX stage register, (ii) exchange by providing result exchange between pipelines by making the Ex-MEM stage register results accessible to the EX stage of the parallel pipeline, and (iii) means for branch pipeline flushing with control hazards by flushing only pipelines that depend on one pipeline that calculates conditions based on branch addresses.

The invention discloses a large language model reasoning acceleration method, device and equipment and a medium, which utilize the difference between a pre-filling stage and a decoding stage in the aspects of calculation characteristics and parallelism to separate the two stages and implement independent batch processing scheduling. Calculating a target pre-filling batch processing value and a target decoding batch processing value which enable the pre-filling time consumption of the pre-filling stage and the decoding time consumption of the decoding stage of the large language model to be closest, and taking the least common multiple of the two as the target batch processing value of the large language model; dividing the plurality of input requests into a plurality of batches according to a target batch processing value, inputting the input requests into a large language model for reasoning, maximizing the system throughput under a hardware constraint condition, aligning the time of two stages, realizing pipeline overlapping on a request level, fully utilizing batch processing computing resources in each period, and improving the efficiency of batch processing. The waiting and idle time between stages is reduced, the overall throughput of the large language model is improved, and then the reasoning speed of the large language model is increased.

The invention discloses a strategy optimization method for cache garbage collection, relates to the technical field of computer cache management, and is used for solving the problem of low cache collection efficiency. According to the method, a cache storage area is abstracted into a plurality of cache partitions, a partition state vector composed of capacity, access and fragmentation degree operation indexes is constructed for each partition, and a strategy evaluation thread periodically reads the partition state vectors to calculate garbage recycling pressure and recycling priority. The upper limit of the number of scanning objects and the upper limit of time slices are distributed for the target cache partition by combining the idle time proportion of the processor, and a recycling execution thread conducts incremental recycling in a batch mode according to a strategy list and records the recycling progress; after recovery, index weight and scanning configuration are adjusted according to release capacity, fragment degree change, hit times and back-to-source request change, so that garbage recovery gives consideration to release space and overhead, the hit rate and the memory utilization rate are improved, and pause and invalid recovery caused by centralized recovery are reduced.

The invention discloses a cloud workflow scheduling method based on dynamic deadline and reliability constraint. The method comprises the following steps: step 1, obtaining a task set of workflow, a workflow deadline and a workflow reliability constraint; 2, calculating an upward sorting value of each task in the task set, and performing descending sorting on the tasks according to the upward sorting values to generate an initial task scheduling sequence; 3, determining a dynamic deadline of the to-be-scheduled task based on the initial task scheduling sequence; 4, determining a target reliability constraint of the to-be-scheduled task based on the initial task scheduling sequence; 5, arranging the virtual machines in a descending order to form a candidate queue; traversing the virtual machines and idle time slots thereof in the candidate queue, and distributing the tasks to be scheduled to the idle time slots; 6, judging whether an unscheduled task exists or not, and if so, returning to the step 3 to repeatedly execute; and if not, adjusting the task scheduling sequence based on a variable neighborhood descent algorithm and a simulated annealing algorithm to complete cloud workflow scheduling.

Methods, systems, and devices for cross temperature collection and reporting in a memory system are described. A memory system may write, to a buffer in response to performing a read operation for a logical address, an indication of the logical address and a read temperature, for the logical address, of the memory system during the read operation. The memory system may determine, in response to detecting an idle time and using the read temperature from the buffer, a cross temperature for the logical address. The memory system may then transmit an indication of the cross temperature for the logical address after determining the cross temperature using the read temperature from the buffer.

The invention provides a demand response bus scheduling method and system considering subway connection, and relates to the technical field of intelligent transportation, and the method comprises the steps: obtaining passenger request data to form a request set; constructing a candidate point library, and providing point location options; acquiring a target subway station timetable, setting a transfer safety margin, establishing a time constraint condition, and matching a subsequent shift if the time constraint condition is not met; the vehicle position, the passenger number, the capacity and the idle time are updated in real time; verifying waiting time, metro connection and vehicle capacity constraints, and generating a feasible scheme set containing point locations, vehicles and estimated time; the waiting time, the walking time, the transfer margin and the vehicle load rate are calculated, an optimal scheme is selected through weighted scoring and fed back to passengers, the scheme is locked after confirmation is received, and an execution basis is provided for vehicle scheduling; and allocating the locking scheme to the corresponding vehicle, and generating a scheduling instruction. According to the invention, efficient configuration of vehicle resources can be realized, and the overall service level of an urban public transport system is improved.