35 results about "Paralleling technique" patented technology

Embodiments of the invention provide data management solutions that go beyond the traditional warehousing system to support advanced analytics. Furthermore, embodiments of the invention relate to systems and methods for extracting data from an existing data warehouse, storing the extracted data in a reusable (intermediate) form using data parallel and compute parallel techniques over cloud, query processing over the data with / without compute parallel techniques, and providing querying using high level querying languages.

The invention provides an OPTICS clustering algorithm based on a Spark big data platform, and relates to a computer information obtaining and processing technology. Parallel data is structurally partitioned, the optimal data set partitioning is obtained, a corresponding RDD is generated, neighbor sample numbers and core distances are calculated in parallel, partitions are subjected to parallel execution of the OPTICS algorithm to obtain a cluster sequence of the partitions, and the cluster sequence is obtained persistently; clusters are given to the partitions according to the cluster sequence, and samples can obtain global cluster numbers by combining the partitions. By means of the Spark distributed parallel technology, the optimal partitioning structure is found, and the cluster sequence of the partitions is obtained through parallel calculation. According to the OPTICS cluster sequence, a user can observe the inherent clustering structure of a data set from different levels of structures, the method can process a large data set which cannot be processed by a serial algorithm, and the time for obtaining the clustering result is greatly shortened.

Embodiments of the invention provide data management solutions that go beyond the traditional warehousing system to support advanced analytics. Furthermore, embodiments of the invention relate to systems and methods for extracting data from an existing data warehouse, storing the extracted data in a reusable (intermediate) form using data parallel and compute parallel techniques over cloud, query processing over the data with / without compute parallel techniques, and providing querying using high level querying languages.

The invention discloses a multi-core parallel method for continuous system simulation based on TBB threading building blocks. The method mainly comprises the following steps: step 1, building a TBB parallel computing platform; step 2, constructing a parallel ParaODES template catalog; step 3, carrying out model initialization; step 4, initializing TBB task scheduling; step 5, judging a simulationtime scope; step 6, calling the parallel ParaODES template; step 7, transferring a simulation result; and step 8, ending TBB task scheduling. The method can realize parallelization processing for continuous system simulation by TBB, improve the simulation calculating speed, meet the requirement of simulation real-time property, and shorten the simulation research period; by parallelization processing with the TBB, the method has simple and flexible process, easy operation and good expansibility; and the TBB adopts a multi-core architecture, which conforms to the development direction from thecurrent parallel technology to a multi-core platform.

An LLC resonant converter includes a first phase with a first primary circuit and a second phase with a second primary circuit. The first primary circuit includes a first shared inductor, and the second primary circuit includes a second shared inductor. The first and second shared inductors are connected in parallel with each other. The first and second primary circuits do not include a capacitor that is connected in parallel with each other.

The invention discloses a method and an electric circuit structure for bit plane quick decoding of EBCOT coefficient in coding and decoding standard of ISO / IEC JPEG2000 image compressing so as to improve JPEG2000 EBCOT decoding efficiency. The position plane decoding electric circuit of the invention can make the original coding block expansion analog pixel to form a new coding block being divided into three bank storage coefficients and coefficient state information, so the invitation capacity of a storage can be reduced, the coefficient can be scanned by using 6x4 window to accelerate the decoding process, simultaneously, a parallel technology is used to further improve the total EBCOT decoding throughput. The invention can improve JPEG2000 image decoding efficiency and can be widely used in digital image product series.

The invention provides an SAS frequency domain signal processing method based on trunking processor. The method adopts parallel technology, multithreading technology and FFTW algorithm, thus omegak algorithm to be transplanted into SAS system and omegak algorithm can be carried out on a trunking processor. The method includes the following steps that: (1) parameter list of SAR system is set; (2) distance direction pulse compression is carried out; (3) azimuth direction non-uniform separate fast Fourier transformation NSFFT is carried out; (4) distance direction and azimuth direction phase compensation and STOLT transformation are carried out; (5) two dimension Fourier inverse transformation is carried out; wherein separate expression of the NSFFT method in the step (3) is showed in the specification. The invention is based on trunking processor and has high cost performance and stability and reliability; OpenMP technology is utilized to realize parallelization, operating rate is high,and operating efficiency is high; and FFTW technology is utilized to realize fast Fourier transformation, operating rate is high, and operating efficiency is high.

An LLC resonant converter includes a first phase with a first primary circuit and a second phase with a second primary circuit. The first primary circuit includes a first shared inductor, and the second primary circuit includes a second shared inductor. The first and second shared inductors are connected in parallel with each other. The first and second primary circuits do not include a capacitor that is connected in parallel with each other.

The invention relates to the technical field of hardware implementation of a self-adaptive filtering algorithm based on LMS (least means square). A self-adapting filter with high performance is designed. The hardware implementation of the LMS self-adapting filter with high performance mainly gives priority to data operation and is based on a delay LMS (DLMS) algorithm. A whole system comprises two modules and two recursive loops, wherein the modules comprise a finite impulse response (FIR) filtration module and a coefficient updating module, and the recursive loops comprise a weight updating loop and an error feedback loop. The FIR filtration module comprises a multiply-accumulate unit unique to the invention, the multiply-accumulate unit is achieved by means of a stream line and a parallel technique, and the multiply-accumulate unit does not vary with the variation of an order when delay occurs. The multiply-accumulate unit optimizes multiplication and addition as a whole, and the multiplication reduces the number of partial products and addition series of an accumulator by utilizing an improved Booth code and 4-2 compressors. A sign function is adopted in the coefficient updating module, and hardware expense required by the module is reduced. According to the hardware implementation of the LMS self-adapting filter with high performance, the self-adapting filter system which is high in speed, small in area and remarkable in filtering effect is achieved.

The invention discloses a multi-source satellite image real-time online processing technical method and device. The computer network technology, the computer parallel technology, the computer multithreading technology and the image processing technology are comprehensively used for multi-source satellite image real-time online processing. The satellite image processing is fast, the computing resource use efficiency is high, and the data packet throughput rate is large. The multi-source satellite image real-time online processing technical method includes that a data packet is received on a high speed network, and the received data packet is organized to be an image; the organized image data are analyzed, and the analyzed image is subjected to storage format conversion and processing; the processed image data are continued to be compressed; and the compressed image data packet is sent back to the high speed network.

The invention provides a seamless automatic image splicing method resistant to subtitle interference. By means of the method, video frames (such as video frames or multiple continuous images shot by a continuously moving or rotating camera) in a varied scene can be continuously spliced to form a complete image. Camera position and pose and scene control point information, proportion and perspective transformation are not required, and rotation of videos (or image sequences) at any angle is not required to be considered, full-automatic color video image seamless splicing is achieved, various high-saliency associated information among the image sequences is fully utilized to perform automatic splicing calculation, the splicing accuracy of adjacent frames can reach to sub pixel, and a parallel technology can be utilized to improve the splicing speed in multiples.

The invention discloses a tactical air navigation system external field beacon simulator, which consists of a frequency synthesizer module, a microwave modulation module and a display control and video processing module, wherein the frequency synthesizer module and the microwave modulation module finish generating and receiving five paths of radio frequency signals on the basis of microwave signalprocessing and a microstrip technology; the display control and video processing module takes a large-scale integrated circuit FPGA (Field Programmable Gate Array) and a high-speed MCU (Single Chip Microcomputer) as the center; and a bell shaped pulse and a PIN (Personal Identification Number) modulation technology are adopted to effectively improve frequency spectrum performance, and the full-frequency-band and full-channel coverage function of the tactical air navigation system external field beacon simulator is realized. A multi-channel parallel technology and the microwave modulation module with a microwave signal processing function are adopted to generate the simulation signals of a plurality of beacon stations, and a one-machine multiple-station function can be realized.

An LLC resonant converter includes a first phase with a first primary circuit and a second phase with a second primary circuit. The first primary circuit includes a first shared inductor, and the second primary circuit includes a second shared inductor. The first and second shared inductors are connected in parallel with each other. The first and second primary circuits do not include a capacitorthat is connected in parallel with each other.

The invention discloses a parallel two-dimensional flow field multivariate data dynamic visual system which comprises the following operation steps: (1) preparing flow field data; (2) computing the particle path; and (3) drawing a flow field diagram. Preparing the flow field data includes a time variable, a latitude and longitude coordinate variable, a speed component variable and a temperature variable and other lattice arrays. Computing the particle path is realized through the Fortran program. Finally the flow field diagram is drawn by using the Python language and the corresponding program package. A curved line vector field is constructed by using a particle tracking method based on the commonly used tool in the field of earth science to more visually describe the two-dimensional flow field of the ocean, and particle tracking and drawing are computed by using the parallel technology so that the processing capacity and the computing speed can be greatly enhanced.

The invention provides a buck DC-DC converter and an electronic device. The buck DC-DC converter comprises an input unit, a first phase buck unit, a second phase buck unit, an output unit and a control unit; the input unit is used for connecting with a direct current source, and the output unit is used for connecting with a direct current load. The first phase buck unit and the second phase buck unit are connected with the direct current load; the first phase buck unit includes a first switch tube; the second phase buck unit includes a second switch tube and a third switch tube; the control unit outputs pulse width modulation signals respectively to the first switch tube, the second switch tube and the third switch tube such that the first phase buck unit and the second phase buck unit outputs a voltage alternately to the output unit. The two-phase alternate parallel technique is used herein, buck capacity of the converter is improved, voltage stress and output current ripple of a switch device are decreased, and power level of the converter is increased.

The invention belongs to the technical field of particle simulation method paralleling, and particularly relates to a body distribution parallel method for eliminating data race in particle simulation algorithm paralleling. The method includes: setting a data race influence range Nc=NsXNsXNs of a network in a three-dimensional particle simulation algorithm, wherein the Ns=2XnMax+1, and the nMax is the maximum value of the data race influence range of the network in the three-dimensional particle simulation algorithm; newly adding a serial number for eliminating a data race method of a subsequent application on all the grids, wherein the numbering rule includes that using grid sets having the length, the width and the height of Ns number as units, successively numbering the units as 1-Ns, circulating this manner till all the grids are traversed; extracting grids with the same serial numbers to be put in a set, and generating Nc sets, wherein all the grids of each set of the Nc sets does not exist data race; and finally serializing the Nc sets. The efficiency of particle simulation algorithm paralleling can be improved through elimination of data race.

An exemplary method includes solving on a computing system an optimal power flow formulation for a plurality of generators in a power system. The solving includes computing using multi-threaded parallelism a plurality of constraints for the formulation, computing using multi-threaded parallelism a plurality of Jacobian functions of the constraints, and computing using multi-threaded parallelism a Hessian of Lagrangian functions. The method further includes outputting results of the solving, wherein the results comprise values of generation levels for the plurality of generators. Apparatus and program products are also disclosed.

The invention belongs to the field of computer vision, and particularly relates to a target tracking algorithm based on kernel correlation filtering and frame difference method fusion. According to the invention, an integrated learning theory is used; aiming at the characteristics of more target tracking pixels, small target and low pixel of satellite video data, a kernel correlation filtering (KCF) algorithm and a frame difference method are selected to be fused, the two tracking algorithms are fused into a strong tracker, and the advantages of the two tracking algorithms are absorbed at thesame time, so that a target tracking task is completed according to the characteristics of the satellite video data. In the multi-algorithm fusion process, the distance between a candidate box and a target is measured mainly by calculating the attraction a of the candidate box, and the time performance of the fusion algorithm is improved by using a parallel technology. According to the invention,tracking can be carried out on satellite video data and a good tracking result can be obtained.

The invention discloses a method and an electric circuit structure for bit plane quick decoding of EBCOT coefficient in coding and decoding standard of ISO / IEC JPEG2000 image compressing so as to improve JPEG2000 EBCOT decoding efficiency. The position plane decoding electric circuit of the invention can make the original coding block expansion analog pixel to form a new coding block being dividedinto three bank storage coefficients and coefficient state information, so the invitation capacity of a storage can be reduced, the coefficient can be scanned by using 6x4 window to accelerate the decoding process, simultaneously, a parallel technology is used to further improve the total EBCOT decoding throughput. The invention can improve JPEG2000 image decoding efficiency and can be widely usedin digital image product series.

The invention provides an SAS frequency domain signal processing method based on trunking processor. The method adopts parallel technology, multithreading technology and FFTW algorithm, thus omegak algorithm to be transplanted into SAS system and omegak algorithm can be carried out on a trunking processor. The method includes the following steps that: (1) parameter list of SAR system is set; (2) distance direction pulse compression is carried out; (3) azimuth direction non-uniform separate fast Fourier transformation NSFFT is carried out; (4) distance direction and azimuth direction phase compensation and STOLT transformation are carried out; (5) two dimension Fourier inverse transformation is carried out; wherein separate expression of the NSFFT method in the step (3) is showed in the specification. The invention is based on trunking processor and has high cost performance and stability and reliability; OpenMP technology is utilized to realize parallelization, operating rate is high, and operating efficiency is high; and FFTW technology is utilized to realize fast Fourier transformation, operating rate is high, and operating efficiency is high.

The invention discloses a spring type parallel bar connector which belongs to the technical field of agricultural machinery. On the upper surface and front side surface of a frame A and a frame B of two parallel operating machines, a set of spring type parallel bars connectors is installed in the chutes of the two frames, one connector is installed on the upper surface of the frame, and the otherconnector is installed on the side surface of the frame. A frame A fixed plate is fixed to the frame A through tightening screws, and parallel radius bars are put in plate locking holes in the frame A fixed plate. The invention can reliably connect two different operating machines, which are driven by one tractor to simultaneously operate, in parallel, enable the two different operating machines to quickly disconnect or lock, and reliably control the relative movement of the two parallel machines, thereby providing reliable technical support for simultaneous processing of multiple operations;the realization of the parallel technology can reduce the number of machine in-field operations and the soil compaction, thereby satisfying the demands for optimal sowing days and reducing the field compaction; and the invention has the advantages of convenient installation and manufacture, simple operation and reliable use.

The invention discloses a multi-core parallel method for continuous system simulation based on TBB threading building blocks. The method mainly comprises the following steps: step 1, building a TBB parallel computing platform; step 2, constructing a parallel ParaODES template catalog; step 3, carrying out model initialization; step 4, initializing TBB task scheduling; step 5, judging a simulationtime scope; step 6, calling the parallel ParaODES template; step 7, transferring a simulation result; and step 8, ending TBB task scheduling. The method can realize parallelization processing for continuous system simulation by TBB, improve the simulation calculating speed, meet the requirement of simulation real-time property, and shorten the simulation research period; by parallelization processing with the TBB, the method has simple and flexible process, easy operation and good expansibility; and the TBB adopts a multi-core architecture, which conforms to the development direction from thecurrent parallel technology to a multi-core platform.

The invention belongs to the technical field of parallel particle simulation methods, in particular to a method for eliminating data competition in parallel operations of particle simulation algorithms. The present invention solves by setting: the data competition influence range of a grid in the one-dimensional particle simulation algorithm Ns=2×nMax+1 grid, and nMax is the maximum value of the data competition influence range of a grid in the one-dimensional particle simulation algorithm ; Then add a new numbering for subsequent application of the method of removing data competition for all grids, the numbering rule takes the grid set with a length of Ns number as a unit, numbering 1-Ns in turn, and looping like this until all grids are traversed Then extract the grids with the same number and put them into a collection to generate Ns collections, each of the Ns collections has no data competition in all the grids; finally, the Ns collections are serialized implement. The invention significantly improves the parallel efficiency of the particle simulation algorithm by eliminating data competition.

The invention belongs to the technical field of remote sensing and geographic information system, and relates to an intelligent method for path optimal selection of parallel processing of remote sensing images based on data segmentation. The invention firstly obtains an input remote sensing image, Then, the information cost o of the parallel processing program , the relative information amount r and the number of elements n of the remote sensing image in each step is calculated automatically according to the training samples;. Then the cost ratio Phi p / Phi 'p and the ratio (rj x ni) / (rj + 1 xni + 1) of relative information and element number of each processing step are calculated. Finally, according to this ratio, the optimal path of parallel processing of remote sensing image is determined and obtained. The invention solves the path dynamic and optimal selection problems caused by the multi-path reachability of the processing model when the parallel technology is applied to the field of distributed storage and processing of remote sensing images, and the principle is simple and practical, and meets the requirements of the shortest time consuming and the highest efficiency of theparallel processing.

The invention discloses a simulation method of a large-scale complex UWB (Ultra Wide Band) channel. The method adopts curvilinear hexahedron to subdivide a complex wireless channel model, and thus modeling at any grid can be achieved. By solving a vector wave equation, the diversity of material can be met, and the distribution of space electric fields can be accurately obtained; obtained electric filed information is deconvolved by a clean algorithm to obtain system impact response and extract the parameter of the channel. The simulation method of the large-scale complex UWB (Ultra Wide Band) channel utilizes the character of a time domain spectrum element method that can solve a matrix equation rapidly, and simultaneously utilizes an efficient parallel technology, so that the analysis on super-large scale wireless channel problems can be obviously accelerated.

The invention belongs to the technical field of parallel particle simulation methods, in particular to a line allocation parallel method for removing data competition in parallel particle simulation algorithms. The present invention solves by setting: the data competition influence range of a grid in the one-dimensional particle simulation algorithm Ns=2×nMax+1 grid, and nMax is the maximum value of the data competition influence range of a grid in the one-dimensional particle simulation algorithm ; Then add a number for subsequent application of the method of removing data competition for all grids. The numbering rule takes the grid set with a length of Ns number as a unit, numbering 1‑Ns in turn, and looping like this until all grids are traversed Then extract the grids with the same number and put them into a collection to generate Ns collections, each of the Ns collections has no data competition in all the grids; finally serialize the Ns collections implement. The invention significantly improves the parallel efficiency of the particle simulation algorithm by eliminating data competition.

The invention belongs to the technical field of parallel particle simulation methods, in particular to a volume allocation parallel method for removing data competition in parallel particle simulation algorithms. The present invention sets: the data competition influence range of a grid in the three-dimensional particle simulation algorithm Nc=Ns×Ns×Ns, wherein Ns=2×nMax+1, and nMax is the data competition influence of a grid in the three-dimensional particle simulation algorithm The maximum value of the range; then add a number to all grids for the subsequent application of the method of removing data competition. The numbering rule takes the grid set whose length, width and height are Ns as the unit, and is numbered 1-Ns in turn, and based on this Loop until all the grids are traversed; then extract the grids with the same number and put them into a set to generate Nc sets; finally execute the Nc sets in series. The invention significantly improves the parallel efficiency of the particle simulation algorithm by eliminating data competition.