148results about How to "Speed up execution" patented technology

The invention relates to a multi-queue multi-priority big data task management system and method for achieving big data task management by utilizing the system. The method comprises the following steps: S1. a client side launches a task request and sets a task queue tag and priority for a task; S2. a resource manager receives the task request launched by the client side, and allocates the task request into the corresponding task queue; S3. each node resource manager reports a resource owned by a distributive node where on which node resource manager is positioned to the corresponding resource manager so as to form a resource pool, and each resource manager allocates the resources for each task queue; and S4. each resource manager sequentially allocates the current task needing to be executed in a plurality of task queues to the node resource managers, and after a task command is received, each node resource manager dispatches a container unit to execute the task. According to the multi-queue multi-priority big data task management system and the method utilizing the system, the unified management and dispatching of the resources can be conducted, the task dispatch execution can be conducted based on multi-queue and multi-priority conditions, the utilization rate of resources can be increased, and the task execution speed can be increased.

The invention relates to the communication field and discloses an authentication method, a network and a system of mobile ad hoc network, which ensures the realization of layer distribution type authentication scheme of cluster mobile ad hoc network; and the requirement on computer processed resources is rational, which can be realized in practical application environment. The invention adopts a group network structure of cluster mobile ad hoc network and advocates layer distribution identity authentication scheme; and the inter-cluster communication used an improved new threshold group signature protocol which improves the original threshold group signature method based on GQ and decreases power index calculation times, improves calculation parallelism and reduces the requirement on processed resources; at the time of initialization, a certificate and a sub key are issued by the system uniformly; cluster heads which have enough numbers of sub keys can jointly recover system keys, thereby carrying out the group signature to issue certificates to new added nodes and using multi-leap serial communication for realizing united group signature.

A storage mamagement method of embedded database,includes that: (1) the storage strategy of data in the memory is that: if logic structure accords with physical structure, the structure is into a divided system area, a main storage database area, a user working area, a log area and a preformed area; the storage strategy of data in the external memory is that: logic structure is divided into a database table, a segment and a database block; physical structure includes a physical file and a physical block; (2) the physical file organization of database is that: a data dictionary is placed at the heading, a user table since then; (3) the memory management of data dictionaryin the external memory is that: the described information of physical file is stored in the file header, denotative definition kind and attribute definition kind adopt a three section type storage form; data dictionary adopt a page type memory management method in the memory; (4) a permanent memory data is stored using T tree and a a permanent external memory data is stored using B+ tree; data and its index are managed using a extended segment memory management method. The invention improves the utilization rate of memory space, and accelerate the speed of data to access.

The inventive IC card includes a first communications unit, a second communications unit, and a passing information management unit accessible to a card application, The passing information management unit retains the processing identifier transmitted from an external R/W via the first communications unit and the PIN transmitted from a cell phone via the second communications unit in association with each other. The card application, receiving a processing identifier and a data request from the external R/W reads the PIN whose corresponding processing identifier matches the received processing identifier from the passing information management unit and transmits the PIN to the external R/W. Thus, even in case a cell phone is in the hands of another malicious person with a PIN written into an IC card, the PIN cannot be read from the IC card unless the processing identifier transmitted from the external R/W matches the processing identifier corresponding to the PIN.

The invention discloses a method for realizing a reconfigurable accelerator customized for a program. The reconfigurable accelerator customized for the program accelerates the program on an FPGA by arranging the FPGA for the prior general-purpose computer system. The method has a main function of analyzing the program, uses functions to calculate information for the runtime of a granularity sampling program, acquires computing-intensive hot spot functions in the program, realizes the hot spot functions as the reconfigurable accelerator on the FPGA, and modifies call of the hot spot functions in the program into call of the corresponding reconfigurable accelerator to accelerate execution of the hot spot functions. The method uses the reconfigurable accelerator to realize the hot spot functions of the program, improves the total speed-up ratio of the program, uses the FPGA to realize the reconfigurable accelerator, achieves the performance of an approximately applied custom integrated circuit, and simultaneously maintains the flexibility of a general-purpose processor.

The invention relates to a cloud computing task scheduling method based on response time optimization. The method includes the steps that a cloud computing system based on response time optimization is constructed, the transmission time of task fragments is calculated, the average processing time of the task fragments at computational nodes is calculated, the total processing time of the task fragments is calculated, task response time is calculated, an objective function of the cloud computing task scheduling problem is constructed, and the task fragmentation scheme of schedulers is calculated. Starting from parallel processing of the task fragments, the cloud computing task scheduling method based on response time optimization uses the longest execution time of task fragmentation as the task response time, the minimum task response time of each scheduler is used for target modeling, and a novel task scheduling method is obtained. The cloud computing task scheduling method based on response time optimization can more reflect the characteristic of task parallel processing. Experiments show that whether the response time value of each scheduler or the objective function value, the cloud computing task scheduling method is superior to both a game playing algorithm and a balance scheduling algorithm, and when the scale of the system is increased or the load is increased, the cloud computing task scheduling method based on response time optimization has obvious advantages compared with the other two algorithms as well.

The invention discloses a hardware accelerating implementation method of a bzip2 compression algorithm, wherein, a hardware accelerator is utilized to implement preposing conversion and stroke length coding which cost a great deal of runtime so as to accelerate the program compression speed. The hardware accelerating implementation method has the characteristics as follows: firstly, an input/output buffer memory of the hardware accelerator is used as a communication interface, and is communicated with a general-purpose computing system through the communication interface; software prepares input data for the hardware accelerator and sorts and reads output data; and so the design of the hardware accelerator is simplified; and secondly, the preposing conversion and stroke length coding is realized in a hardware manner, a fully expanding 2048-bit parallel comparator and a shifter are adopted, so that the program execution is accelerated, the data compression speed of the bzip2 algorithm is accelerated, and the program performance is enhanced effectively.

The invention enables appropriately concentrating instruction strings or data pieces sporadically present in a plurality of regions over more than one compilation unit and adjusting the front-and-rear relationship of executed instruction strings without changing the program compilation unit such as a file, subroutine, or function and also without creating a link processing program for batch processing of the system as a whole. Different section names are given to the executed instruction strings and the unexecuted instruction strings and the referenced data and the unreferenced data of an object program respectively, so that, when an execution module is generated from the object program by linking, the sections having an executed section name and the sections having an unexecuted section name in a plurality of files may be aggregated respectively to divide the instructions into an execution portion and an unexecution portion.

The invention relates to a method for realizing a rear-end translation framework of a heterogeneous multi-core virtual execution environment based on CPU-GPU (Central Processing Unit-Graphic Processing Unit) in the technical field of computers. The method comprises the following steps of: carrying out dynamic translation processing based on a basic block on binary codes so as to obtain a middle representation code; then, analyzing and labeling a cyclic border, a cyclic index and a temporary variable in the generated middle representation code; recognizing access operation in a program by filtering an LD (Laser Disc) and a storage instrument in the middle representation code, recognizing a variable type according to an input cycle information table, labeling the variable according to the corresponding variable type and recording the label in the middle representation code; separating a loop body from the labeled middle representation code through the judgment of a cyclic condition and the branch jump and extracting the operational part of the loop body; and finally, translating the middle representation code of the extracted operational part into a PTX (Performance Toolbox) code to realize the rear-end translation framework of the multi-core virtual execution environment. The invention is beneficial to promoting the wide application of the general operation of the GPU.

The invention discloses a photoelectrical fire control system and method of an armed unmanned aerial vehicle. The photoelectrical fire control system comprises an unmanned aerial vehicle body, an electric source, an electrical pan tilt, a camera, a distance meter, a sensor, a computer, a firing angle adjusting mechanism, a communication module and a ground display and control platform, wherein theelectrical pan tilt controls the camera and the laser distance meter to catch, track and measure a target; the computer calculates firing elements according to received parameters by a solving hit method; and the adjusting mechanism and the unmanned aerial vehicle are used for executing the command to aim at the target. The photoelectrical fire control system disclosed by the invention has the beneficial effects of being compact in structure, small in size, high in accuracy and high in mechanism executing speed; the specific requirements for being light in weighting, and automatically and quickly aiming at and hitting the target of the fire control system of the unmanned aerial vehicle are well met; the working flow of the fire control system and the solving hit method are low in occupiedresources, and high in solving efficiency; and the requirements of the armed unmanned aerial vehicle for accurately measuring the target, precisely aiming at the target and precisely hitting the target are met.

A Very Long Instruction Word (VLIW) processor having an instruction set with a reduced size resulting in a small number of bits being necessary to specify registers. The VLIW processor includes a register file, and first through third operation units, and executes a very long instruction word. Further, the very long instruction word includes a register specifying field which specifies a least one of the registers in the register file and a plurality of instructions. The operand of each instruction includes bits src1, src2, and dst, which indicate whether or not the registers specified by the register specifying field are to be used as the source register and the destination register.

The invention discloses an operating method for decreasing read delay of a solid-state disk; the operating method is characterized in that firmware issues a conventional operation command queue to an NAND controller through a CPU (central processing unit), the NAND controller reads a command from the command queue via an arbitration mechanism and transmits the command to NAND flash, data are transmitted to or read from the NAND flash through a data path, NAND medium state and NAND controller state for page write operation or page erase operation transmitted by the firmware previously to the command queue are fed back to the CPU, and a read operation is transmitted to the NAND controller through Bypass command. By combining firmware and hardware, the execution of the read operation is accelerated in order to solve the problem that NAND solid-state storage equipment has too long delay in data reading response.

The invention provides a method and system for reconstructing a three-dimensional image of an electronic speculum. The method for reconstructing the three-dimensional image of the electronic speculum comprises the following steps: a central processing unit (CPU) divides an object in an image to be subjected to iteration reconstruction into slices, and constructs a frame that the CPU cooperatively calculates with a many integrated core (MIC) processor; the CPU schedules all equipment in the frame to obtain reconstruction image values and measurement projection image values of respective current slices in a multi-threading parallel mode, and calculates final reconstruction image values of the current slices through a re-projection algorithm and a back projection algorithm according to the reconstruction image values and the measurement projection image values in sequence till calculating the final reconstruction image values of all slices. According to the method and the system for implementing the electronic speculum three-dimensional image reconstruction, through the cooperative calculation of the CPU and the MIC, the execution of an iteration reconstruction method is greatly accelerated, so that image reconstruction tasks can be rapidly accomplished.

The invention provides a fluid-analysis-based fluid parameter estimation method. The method comprises the steps: carrying out empirical mode decomposition; to be specific, decomposing each frame of collected fluid velocity field into modes with different frequencies, thereby keeping information in a frequency domain; carrying out fluid velocity field dimensionality reduction; to be specific, carrying out integration on the decomposed model set and compressing the integrated one into a base and projecting calculation in three-dimensional space to sub space with a small scale, thereby reducing the calculation cost substantially; carrying out fluid simulation parameter estimation; to be specific, carrying out fluid simulation parameter back deduction in the sub space by using the collected fluid data according to an euler fluid simulation method; and carrying out re-simulation or fluid sequence editing; to be specific, reproducing a collected fluid sequence by using the fluid simulation parameter obtained by back deduction or carrying out parameter and boundary condition modification, and editing the fluid sequence. According to the invention, the empirical mode decomposition process is accelerated by using a GPU; and implementation of the process in the three-dimensional space is improved, so that the process can be completed within acceptable time.

An RF transceiver includes an Envelope Restoration (ER) transmitter (TX) and a receiver (RX). A method includes providing the TX with at least one programmable delay element in at least one of an AM path and a PM path; making an RF connection between an output of the TX and an input of the RX; and responsive to an output of the RX when receiving a signal through the RF connection, determining at least one delay value for use in programming the at least one programmable delay element. Making an RF connection includes measuring an effect of a delay mismatch between the AM path and the PM path for use in determining the at least one delay value. In one embodiment measuring performs a power measurement such as an Adjacent Channel Leakage Ratio (ACLR) power measurement, the delay value is determined as being a value that minimizes the ACLR, and the RX is tuned, when receiving a signal through the RF connection, to an RX carrier frequency that is about one channel spacing away from a TX carrier frequency. In another embodiment measuring performs an Own-Channel Power (OCP) power measurement, the delay value is determined as being a vale that maximizes the OCP, and the RX is tuned to an RX carrier frequency that is substantially equal to a TX carrier frequency. In another embodiment measuring performs a signal quality measurement, such as a Bit Error Ratio (BER) measurement, the delay value is determined as being a value that minimizes the BER, and the RX is tuned to an RX carrier frequency that is substantially equal to a TX carrier frequency.

The invention discloses a text classification method and device based on a neural network model, a terminal and a storage medium, and the method comprises the steps: searching a learning rate according to the gradient of an optimizer based on a random gradient descent and momentum method in the deep learning training process of the text classification model, and then improving the attenuation ratein the optimizer according to the reduction ratio of the learning rate; Based on the improved attenuation rate, the searched learning rate and the gradient, optimizing the neural network model through an optimizer until the model completes optimization training; And finally, inputting the text word vector corresponding to the target text into the neural network model subjected to optimization training to obtain a text classification result. In the text classification model learning and training process, a neural network model is optimized by combining a momentum method, random gradient descent and dynamic adjustment of the learning rate and the attenuation rate in an optimizer, the convergence speed and the execution effect of the model are helped to be increased, and then the accuracy ofa classification result during text classification is improved.

The invention discloses a health management method based on a large data intelligent algorithm and system. The health management method comprises steps: based on evidence-based medicine, epidemiologyand crowd intervention data, a multi-layer cumulative logistic regression model is established, multi-dimensional individual health measurement indexes and a nutrition and health intervention scheme library are generated, an individual nutrition and health intervention scheme decision tree model is established by using the individual health measurement index information, and an individual nutrition and health intervention scheme is acquired. During the nutrition and health intervention scheme execution process, health assessment is carried out regularly and best intervention effects are achieved. The management system established based on the method is composed of three modules: a personal health information acquisition module, a calculation module and a display and supervision module. Through the health management method and the system, scientific, personalized and traceable health management solutions can be provided for residents.

The invention discloses a method and a system for enhancing digital image resolution to solve the problems that the further processing and the analysis of the image are influenced due to the low resolution of the prior video image. The method comprises the following steps: image interpolation, namely carrying out nearest-neighbor interpolation for a brightness map of a low-resolution image, and improving a Bayesian image interpolation model; and interpolation correction, namely correcting a brightness map of a high-resolution image by using the improved Bayesian image interpolation model. Aiming at a color image, the method provides image pre-processing and post-processing, wherein the image pre-processing carries out color space conversion for the low-resolution color image, and the imagepost-processing reversely converts the corrected high-resolution brightness map and a chrominance component map into the color images. Aiming at the problems that the prior Bayesian interpolation method has large calculated amount and low execution speed, the execution of interpolation is accelerated by quickly determining the falling direction and step length; meanwhile, the method and the system have edge keeping capacity, and can effectively recover image details and improve the solution of the images.

The invention discloses a method for accelerating a network-based adaptive reconfiguring processing unit array system. According to the method, an adaptive reconfiguring processing unit network is formed by computing nodes formed by general processors and reconfiguring equipment. A program needing to be executed is divided into a collection of tasks which can operate independently; and each task in the collection of the tasks is allocated through the adaptive reconfiguring processing unit network to find a suitable computing resource to be executed and can be reconfigured by making use of the characteristics of high efficiency and flexibility of the reconfiguring equipment, so that the function of the reconfiguring equipment can meet the demands of different tasks. Under the action of the adaptive reconfiguring processing unit network, the network-based adaptive reconfiguring processing unit array system can intelligently allocate the tasks according to the condition of nodes in the network and primarily allocates the tasks to other nodes with idle computing resources so as to reduce the time of executing the program, improve the utilization rate of the reconfiguring equipment and achieve the purpose of accelerating the program.

A method to speed up the execution of repeatable commands and a microcontroller able to speed up the execution of repeatable commands are disclosed. When the microcontroller is to execute repeatable commands in a program, it temporarily stores repeatable commands to a storage unit. If the execution of the repeatable command loop continues, then the repeatable command loop is retrieved from the storage unit and executed at higher clock cycle frequency. At the start and end of the repeatable command loop are respectively defined by a starting point and an end point for determining whether the repeatable command loop should continue to execute. The microcontroller thereby speeds up the execution of the repeatable command and the performance thereof.

The invention discloses a public sentiment handling method and device based on television reports. The public sentiment handling method and device are used for performing timely and effective negative crisis handling to improve the influence of positive users when negative public sentiment report information appears in the television reports. The method includes the steps that according to a keyword set by a user, television report information related to the keyword is monitored, wherein the keyword is a user name; if the television report information related to the keyword is monitored, whether the television report information is negative public sentiment information is judged; if yes, the negative influence parameter of the negative public sentiment information is calculated.

The embodiment of the invention discloses a service processing method and an information processing method of a block chain network, and node equipment. The service processing method comprises the steps: obtaining routing information configured for a target service node by a management node when the target service node receives a service request, wherein the routing information comprises a corresponding relationship between a service type and a processing node, the processing node is used for processing a service indicated by the service type, and the processing node is a node except a targetservice node and a management node in the block chain network; obtaining a target processing node corresponding to a target service type from the routing information according to the target service type to which the service indicated by the service request belongs; and sending the service request to the target processing node to enable the target processing node to execute the service indicated bythe service request. By adopting the embodiment of the invention, the routing information of the service node is uniformly distributed by the management node, invalid routing is avoided, and the service execution efficiency is improved.