127results about How to "Short execution time" patented technology

The present invention relates to a fine-granularity files and directory edition management method based on snapshots, belonging to the multi-version document system field. The present invention separates a name space consisting of files and dirnames in a whole file system from an edition space representing the generating periods of different editions, and adopts relatively independent strategies to execute management, forming a hierarchical two-dimensional structure, i.e. forming a hierarchical structure from a root directory to a file in the name space; in the edition space, the editions of files and directory are organized through an index structure chronologically, forming a hierarchical structure in the edition space. The retrieval of the name space adopts an index strategy based on dynamic hash. The retrieval of the edition space adopts an index strategy based on a red-black tree. The directory edition and file edition respectively adopt a red-black tree structure variant aiming at the respective characteristics. The present invention can greatly improve the usability and the performance of the system, and controls the amount of consumption of time and space resulting from the maintenance of historical editions in an acceptable scope.

The present invention presents a number of improvements for managing erase processes in non-volatile memory. Such memory systems typically manage the memory by logically organize the basic unit of physical erase (erase block) into composite logical groupings (meta-blocks or logical group), where an erase block generally consists of a number of sectors. When an erase command is received, the specified sectors are checked against the memory system's control data. If the specified sectors span any full logical grouping, the full logical groupings can each be treated a whole and erased according to one process (such as performing a true, physical erase), while other sectors are “logically” erased at the sector level by standard techniques.

The invention discloses a quantum-behaved particle swarm optimization (QPSO) based multi-objective dynamic workflow scheduling method, and belongs to the technical field of cloud computing. The method includes the steps: inputting a workflow and a QoS (quality of service) request; acquiring state information of virtual machines and transmission information among the virtual machines; setting a to-be-executed task set V', and setting objective functions of time, cost and reliability for a task schedule in the V'; allocating optimal resources to the to-be-executed tasks by the aid of QPSO, and judging whether total time, total cost and total reliability of task execution meet the QoS request of a user or not after the tasks are executed; dynamically updating the V', transmission speed among the virtual machines and operating speeds of the virtual machines. By means of dynamically partitioning the workflow and dynamically updating network bandwidth information, the optimal resources are allocated to the workflow tasks accurately, errors between the calculated time and actual execution time and the calculated cost and actual execution cost are reduced, time can be shortened, and cost is reduced while reliability is enhanced.

The invention discloses an ETL (Extraction-Transformation-Loading) dispatching system and method for error-correction restarting and automatic-judgment starting. The ETL dispatching system structurally comprises a business system database server, a data warehouse server, an ETL dispatching and monitoring server and an ETL dispatching platform server. The ETL dispatching method disclosed by the invention comprises the following steps of: dividing a data warehouse ETL task into a plurality of subject tasks, wherein each subject task comprises a plurality of sub-tasks; configuring information according to a preset executing sequence and automatically judging the end time for successfully executing the pre-posed associated sub-tasks; executing follow-up sub-tasks, stopping the execution of the associated follow-up sub-tasks according to the sub-task correlation after the execution of the sub-tasks has exception faults, and recording the exceptional sub-tasks; and only processing the exceptional sub-tasks and the associated follow-up sub-tasks in a way of automatically restarting for many times. According to the ETL dispatching system and method disclosed by the invention, the problem that the follow-up sub-tasks are continuously executed when the pre-posed sub-tasks have faults in the ETL task dispatching is solved; the task executing efficiency is improved; the manual intervention is reduced; and the operation loss is reduced.

The invention discloses an indoor positioning method based on MEMS insert device and android smart mobile phone fusion, and belongs to the field of data processing and mobile terminal application. The method is characterized in that a Kalman filtering-based zero speed correction accumulative error elimination algorithm is adopted based on the motion characteristics of indoor positioning pedestrians and the drift error of an MEMS inert device on the premise of MEMS insert device and android smart mobile phone fusion; and different data sampling modes are designed based on the advantages and disadvantages of a gyroscope and an electron compass and the motion characteristics of the pedestrians in the pedestrian motion course angle determination process, and are fused with a particle filtering course angle determination algorithm in order to further increase the pedestrian course precision. The method can reduce the accumulative error in the inertial navigation resolving process, and allows accurate pedestrian displacement information and course angle information to be obtained.

The present invention presents a number of improvements for managing erase processes in non-volatile memory. Such memory systems typically manage the memory by logically organize the basic unit of physical erase (erase block) into composite logical groupings (meta-blocks or logical group), where an erase block generally consists of a number of sectors. When an erase command is received, the specified sectors are checked against the memory system's control data. If the specified sectors span any full logical grouping, the full logical groupings can each be treated a whole and erased according to one process (such as performing a true, physical erase), while other sectors are “logically” erased at the sector level by standard techniques.

An integrated circuit is provided comprising a plurality of processing modules (M, S) and a network (N) arranged for coupling said processing modules (M, S). Said integrated circuit comprises a first processing module (M) for encoding an atomic operation into a first transaction and for issuing said first transaction to at least one second processing module (S) . In addition, a transaction decoding means (TDM) for decoding the issued first transaction into at least one second transaction is provided.

Techniques and structures are disclosed for a processor supporting checkpointing to operate effectively in scouting mode while a maximum number of supported checkpoints are active. Operation in scouting mode may include using bypass logic and a set of register storage locations to store and / or forward in-flight instruction results that were calculated during scouting mode. These forwarded results may be used during scouting mode to calculate memory load addresses for yet other in-flight instructions, and the processor may accordingly cause data to be prefetched from these calculated memory load addresses. The set of register storage locations may comprise a working register file or an active portion of a multiported register file.

The invention relates to a distributed type multi-robot synchronous swarming control method. The method comprises the steps of: identifying not more than six key neighbor robots in all the neighbor robots with the current position of a robot as a center, the detection distance of a sensor as a radius, and other robots in the circular area as neighbor robots, and designing a position control operator of the current robot by using an attraction-repulsion function among the current robot and the neighbor robots; designing a speed direction control operator of the current robot according to the linearization expression of a Vicsel model; and weighing and integrating the position control operator and the speed direction control operator to form a synchronous swarming control rate of the current robot. The invention visually demonstrates network topology and control action relations among robot positions, breaks the restriction on the scale size of robots, realizes purely distributed type extensible coordination control, guarantees communication of network topology structures in the process of group evolution, and can adjust parameters to avoid collision when the scale of the neighbor robots increases continuously and realize self-adapting to the change of the position relations of the robots.

The invention discloses an artificial intelligence operation method and device applied to games; the method comprises the following steps: receiving status information sent by a client end, wherein the status information is used for indicating present game status of the client end; using a built model to obtain motion indication information, wherein the motion indication information is used for indicating the next motion of the present game status, the model is obtained by training a plurality of motion combinations in the client end, and each motion combination comprises a plurality of motions executed in sequence; sending the motion indication information to the client end. The method and device can solve the technical problems that an artificial intelligence operation algorithm in the prior art is poor in game operation strategy programming capability.

A method and arrangement for using a resource of a hardware platform with at least two virtual machines, wherein a virtualization controller for administering the virtual machines executes on the hardware platform, a first machine of the virtual machines with a first operating system executes separately on a first processor core or a first processor, and further virtual machines execute on a further processor core or processor. One of the further virtual machines comprises a service machine including a first driver for accessing the resource, where the first operating system includes a para-virtualized second driver for accessing the resource, and requests by the para-virtualized second driver to access the resource are directed to the first driver that converts the requests into access operations to the resource.

The present invention provides an operational processing apparatus which can guarantee a period for executing instructions in the shortest cycle when the operational processing apparatus synchronizes with a hardware accelerator. A processor in the present invention simultaneously issues and executes instructions including instruction groups having a simultaneously issueable instruction. The processor executes a program including a specific instruction. The specific instruction instructs to exclude an instruction subsequent to the specific instruction out of the instruction groups including the specific instruction, and to suspend issuing the instruction subsequent to the specific instruction only during a predetermined period immediately after the specific instruction is issued.

An intrusion-prevention enclosure, of the type which comprises a plurality of modular panels, which can be mutually associated in an orderly and predefined manner so as to provide a barrier structure, and comprising a plurality of ground supporting uprights provided, substantially proximate to a respective upper end, with detachable engagement elements, which can engage in openings of the panels, for the facilitated assembly of the enclosure by a single operator.

The embodiment of the invention provides a memory leak detecting method and a memory leak detecting device. A memory comprises a doubly linked list of an allocated memory; and the doubly linked list comprises memory header information of one or more allocated memory blocks. The method comprises the steps of: determining a detection time interval; traversing the header information in the detection time interval in the doubly linked list; and in accordance with the memory header information, determining whether the corresponding one or more allocated memory blocks are subject to memory leak. By the memory leak detecting method and the memory leak detecting device, memory leak can be visually detected in real time, the software performance is improved, and manual operation is reduced or avoided, so that the method and the device are labor-saving and time-saving, the operation difficulty is reduced, the working efficiency is improved, the cost is reduced, the accuracy of the detection of memory leak is improved, and the leaked memory can be recycled in the entity life cycle.

The embodiment of the invention provides a method, a system and equipment for processing a circular task. The method comprises the following steps: a configuration information register module receives configuration information from a bus, and a circular controller module obtains the configuration information from the configuration information register module; an input FIFO module reads data according to the configuration information of an input FIFO; the input selecting module writes the data in the input FIFO into a specific position of an RC input register set according to the configuration information of a calculation unit; a reconfigurable unit array executes a circular task according to the configuration information of the calculation unit and generates a calculation result, namely the output of an RC; an output selecting module writes the output of the RC into an output FIFO in a specific sequence according to the configuration information of the calculation unit; and the output FIFO module transmits the data according to the configuration information of the output FIFO. The scheme provided by the embodiment of the invention solves the problem of longer executing time of the circular task on a processor, improves the running speed of the circular task and reduces the total running time of an application program.

A semiconductor device includes a FIFO, a test data write circuit that sequentially writes a plurality of test data to the FIFO in synchronization with a first clock signal, and a test control circuit that, in parallel with writing of the plurality of test data to the FIFO by the test data write circuit, sequentially reads a plurality of test data stored in the FIFO in synchronization with a second clock signal that is not synchronous with the first clock signal and performs a scan test of a circuit to be tested.

The invention provides a computer and a working method thereof. The computer system comprises a central processor and a memory, wherein the memory is provided with a resource distribution management storage region for storing an execution sequence when the central processor executes application programs; the memory is provided with a resource distribution management unit for dynamically changing the execution sequence according to the resource request quantity occupied by a plurality of application programs; and the execution sequence is as follows: the execution priority of the application program with small resource request quantity is higher than that with large resource request quantity. The method comprises the following steps: the resource distribution management unit receives resource request quantity information transmitted by the plurality of application programs, and writes the execution sequence of the plurality application programs into the resource distribution managementstorage region; the resource distribution management unit dynamically changes the execution sequence thereof according to the resource request quantity occupied by the application programs; and the central processor executes the application programs according to the execution sequence. The computer system and the method can improve the response speed of the application program with small resourcerequest quantity to avoid generation of system halted phenomenon.

The invention discloses a distributed erasure code storage system data restoration method based on simulated annealing. The method comprises the following steps: constructing a distributed storage system; the monitoring node judging that the storage node has a fault; the monitoring node selecting a storage node as a repair node to repair the data and store the lost data; the repair node obtainingthe network transmission speed; the repair node obtaining the erasure code generation matrix, the failure node number and the current node number and obtaining a repair scheme through a search algorithm; the repair node constructing a data reading request according to the repair scheme and sending the data reading request to each storage node; each complete storage node reading corresponding element data and returns the element data to the repair node; the restoration node executing decoding operation to reconstruct the lost data; and the restoration node feeding back the metadata informationto the monitoring node and storing the data locally. The bandwidth occupation of data restoration of the CRS erasure codes is effectively reduced, the adverse influence of data restoration on the system is reduced, and the data restoration speed is increased.

The invention relates to a storage system reference evaluation program generating method and apparatus and belongs to the high-performance computing performance evaluation field. The storage system reference evaluation program generating method includes the following steps that: intermediate codes are obtained from a standardized source program; slicing starting points of the intermediate codes are identified; static program slicing is carried out based on identified slicing starting points; and simplified codes can be generated based on obtained static program slices. The method is based program slicing technology; data dependence, control dependence and communication dependence are analyzed in a compiler; all statement operation related to input and output is automatically extracted so as to form the compilable, executable, manually-readable and transplantable simplified program codes; the simplified program can faithfully reflect input and output load characteristics of an original application program, and less time and fewer resources are needed to execute the simplified program, and therefore, rapid evaluation can be facilitated. The simplified program generating method is very effective in fast evaluation of the performance of a high-performance storage system and design assistance of the storage system.

The invention relates to a sensorless initial position detecting method for a permanent magnet synchronous motor. The sensorless initial position detecting method comprises the steps of: adopting a virtual pulsating high-frequency injection method for calculating a rotor position, injecting a high-frequency voltage into a virtual direct axis, establishing a virtual rotating coordinate system, utilizing the virtual rotating coordinate system and a mathematical model of a permanent magnet synchronous motor coordinate system to calculate straight-axis and quadrature-axis currents, multiplying a straight-axis current signal and a quadrature-axis current signal by a voltage frequency containing an injection signal and a cosine value of an assumed rotational frequency simultaneously and separately, subjecting the results to low-pass filtering, and deriving out arc tangent of the two currents after filtering to obtain the rotor position preliminarily; and utilizing a sine and cosine relationship between the straight-axis and quadrature-axis currents to further obtain the rotor position, adopting a carrier frequency component method to replace a secondary injection method for conducting magnetic pole judgment, and correcting N-S poles to obtain the actual rotor position. The sensorless initial position detecting method has the advantages of high precision and simple implementation, and can be used for improving the sensorless starting control precision of the permanent magnet synchronous motor used for fans, pumps, compressors or electric vehicles.

The invention discloses an improved Kmeans clustering method based on SimHash. The method includes the following steps: 1. using the SimHash algorithm to compute the fingerprint of each to-be-clustered file, generating one SimHash table; 2. selecting one to-be-clustered file, based on the fingerprint of the file, checking the SimHash table to obtain one similar file set S0; 3. putting the files from the S0 that have an Overlap value with the to-be-clustered file greater than a threshold [alpha] to a similar file S1; 4. back to a cluster C0 belonged to the S1, computing the distances between the to-be-clustered file and the centers of mass of all clusters in the C0, classifying the to-be-clustered file to a cluster I which is closest to the to-be-clustered file and smaller than the threshold and updating the center of mass of the cluster I, or creating a new cluster k, and taking the to-be-clustered file as a first element of the cluster k; if C0 is empty, creating a cluster j, taking the to-be-filed cluster as a first element of the cluster j. According to the invention, the method increases the effects of clustering.

The invention provides a personnel indoor positioning method based on an adaptive SIFI (scale invariant feature transform) algorithm. The method is characterized by: firstly, carrying out feature extraction on an global motion image sequence; then, taking a correct matching point as a sample set to estimate a camera motion model so as to obtain an actual offset of a camera. In the method of the invention, a Lagrange parabola interpolation is introduced. Last three times of model matching results are used to predict an overlapping region of a reference frame image and a current frame image. On the overlapping region, feature points are extracted and feature matching is performed so that a lot of information redundancies in a video image sequence can be eliminated, a processing speed of the each frame image can be accelerated, the validity of the feature points to be matched can be increased and mismatching can be reduced. Therefore, the algorithm is accurate and real-time and can be used for the indoor personnel positioning system.

The invention provides an indoor positioning and navigation method based on a MPU6050 and a smart phone. The MPU6050 wirelessly transmits double-feet walking measurement data of a user to the smart phone through Bluetooth; the position error of an inertia device itself is reduced through a zero-bias correction algorithm, and with a modified zero velocity correction accumulated error reducing algorithm, a data collected by an acceleration sensor is subjected to time-domain integration for one time to acquire speed, and the speed is then corrected through a modified zero velocity detection method; then frequency domain integration is carried out for one time to calculate walking displacement information which is more accurate; in determination of pedestrian movement direction angle, on the basis of advantages and defects of electronic compass and gyroscope on the smart phone, sliding mean wave filtering and angle compensation aslgorithms are designed; through setting of a threshold value, a proper direction is selected, thus increasing the accuracy of the direction of a pedestrian. By fusing the data of both feet, the weight of left and right feet is obtained through training by neural network, thereby obtaining more accurate pedestrian track information and finally acquiring more accurate pedestrian position information.

The present invention discloses a method to obtain user equipment position information, which includes procedures below: A. A network control equipment stimulates a user equipment not in a special channel to send an uplink signaling to a base station; B. The user equipment sends an uplink signaling to the base station. The base station measures an angle of arrival corresponding to the uplink signaling. The technic proposal of the present invention enables the base station to measure the angle of arrival corresponding to the uplink signaling sent by the UE by stimulating the UE not in the special channel to send the uplink signaling, thus obtaining position of the UE not in the special channel through a network. In addition, the technic proposal obtains the position information of UE not in the special channel without creating a special channel to save network resource. The method of obtaining UE position information needs a short execution period and simple flows.

An embodiment of the present invention provides a system and method for medical image processing. The proposed system includes a grid computing framework adapted for receiving patient data including one or more patient-scan images from an end-user application, and for scheduling image processing tasks to a plurality of nodes of a grid computing network. Each of the nodes includes a central processing unit and at least one of the nodes includes programmable graphics processing unit hardware. The proposed system further includes a second framework for image processing using graphics processing unit that is operative on each node of the network. The second framework operative on any node is adapted to execute the image processing task scheduled to that node based upon the availability of graphics processing unit hardware in that node. When graphics processing unit hardware is available in the node, the second framework is adapted to execute the task on the graphics processing unit of the node using stream computation. When graphics processing unit hardware is not available in the node, the second framework is adapted to execute the task on the central processing unit of the node.

The invention provides a control method of an S-shaped curve speed with a non-zero initial accelerated speed. The method is decomposed into three levels, wherein a speed planning level is used for carrying out speed changing planning according to given parameters to obtain increased acceleration stopping time, uniform acceleration stopping time, decreased acceleration stopping time and a current increased accelerated speed in a speed changing process; a stepping planning level is used for calculating positions, speeds and accelerated speeds at any moment according to results of the speed planning level; a position planning level is used for calculating next-step movement position, speed and accelerated speed of a current starting point according to calculation results of the speed planning level and the stepping planning level, and guaranteeing that the speed is not greater than a terminal point speed when an object finally moves to a terminal point; and when a planned length of a current section chain is not enough, the current section chain is connected with the next section chain, so that the length of the section chain can be planned to be enough. In each step of speed planning process, a maximum speed predication step is operated and planning is carried out in time in order to reach the terminal point speed; and algorithm can be rapidly realized and hardware requirements and cost are reduced.

An optical add / drop multiplexer including one or more optical drop multiplexers connected in free space or fused by optical fiber pigtails, a wavelength blocker with an input port connected to an output port of the optical drop multiplexer through the fusion of the fiber pigtails, one or more optical add multiplexers connected in free space or fused by fiber pigtails, a digital signal processor, an analog-to-digital signal converter, a digital-to-analog converter, and a plurality of electronic control and feedback loops for tuning and scanning an optical wavelength.

The invention provides a method for scheduling and configuring resources of associated task flows in a cloud. The method includes: according to the set weight of task parameters, determining the backup priority for each subtask in the associated task flow, and then reconstructing the associated task flow according to the preset backup model, looking for local critical paths in the reconstructed associated task flow model, assigning computing resources with the best performance to the sub-tasks with the highest backup priority, considering the relationship between the execution cost of associated task flow and computing resources, taking the execution time of associated task flow as the main optimization objective, the execution cost of backup strategy can be reduced as much as possible onthe basis of the shortest execution time of tasks.

A method is disclosed for operating a rotating electric machine, wherein the rotating electric machine is connected by phases to a converter circuit having a direct-current circuit for switching at least two voltage levels, and the phases of the converter circuit are connected to the direct-current circuit according to a selected switch state combination of switch states of power semi-conductor switches of the converter circuit. Exemplary embodiments can reduce the switching frequency of the power semi-conductor switches, based on a prediction of further behavior of the overall system, and selection of an optimum switching state combination.