92 results about "Parallel sorting" patented technology

The present invention reveals a sorting system for use in manual sorting, which presents a detached ephemeral display moving in a manner corresponding to the movement of the article, by which an article to be sorted can be quickly and easily identified. To accomplish its purpose, the device comprises: feed conveyors; a switching unit; optical readers positioned to capture destination indicia affixed to each article; a detached moving display which remains close to the article to be sorted and presents information representative of the article's destination location; a destination location which signals when a related article is approaching; and a controller capable of assigning destination locations and controlling display devices.

A system and method are provided for coordinating concurrent processing of a plurality of messages communicated over a network between a source system and a target system. The plurality of messages are configured for including a pair of related messages having a common first unique message identifier and at least one unrelated message having a second unique message identifier different from the first unique identifier. The system and method comprise an input queue for storing the plurality of messages when received and an output message queue for storing the plurality of messages during message processing. The output message queue includes a plurality of execution streams for facilitating concurrent processing of the at least one unrelated message with at least one of the pair of related messages. The system and method also include a sequencer module coupled to the output message queue for determining using the first and second unique identifiers which of the plurality of messages are the pair of related messages and which of the plurality of messages are the at least one unrelated message. The sequencer module is further configured for identifying a sequence order for the pair of related messages according to a respective sequence indicator associated with each of the related messages, such that the sequence indicators are configured for use in assigning a first position in the sequence order for a first message of the pair of related messages and for use in assigning a second position in the sequence order for a second message of the pair of related messages. The system and method also have a registry coupled to the sequencer module and configured for storing a pending message status indicator for the first message in the output message queue, wherein the sequencer module inhibits the progression of processing through the output message queue for the second message until the pending message status for the first message is removed from the registry while facilitating concurrent processing of the at least one unrelated message though the output message queue.

The invention discloses a data parallel sequencing method and system. The system comprises a data source, a plurality of parallel processing units connected with the data source through a network and a communication interface. The method comprises the steps that data to be sequenced are divided into a plurality of data blocks, and the parallel processing units respectively obtain the data blocks and conduct sampling; a first parallel processing unit summarizes and sequences sampled data of the parallel processing units, the global sequencing interval sequence is determined according to the parallel processing units, and data intervals in the global sequencing interval sequence sequentially correspond to the parallel processing units; the parallel processing units judge the data intervals to which the data in the data blocks obtained by the parallel processing units belong, and the data are distributed to corresponding parallel processing units; the parallel processing units receive the data and conduct global sequencing; global sequencing results of the parallel processing units are combined in a sequencing mode. The data parallel sequencing method and system improve sequencing speed of large-scale data and meanwhile have high expansibility to data volume.

The invention relates to a method and device for detecting large-scale social network communities. The method includes the steps that input large-scale social networks are modeled into a map G = (V, E); all nodes on the map G are sequenced in a descending order according to the size relation of node degrees through a parallel sorting algorithm, and the sum DSum of effective degrees of all the nodes on the map G is calculated; DSum/P serves as an equally-dividing benchmark reference value, and the map G is equally divided into P sub maps through a load balancing method; the P sub maps are traversed for looking for triangles on the map G through a MapReduce parallel computing model, parallel multilayer coarsening is conducted on the map G based on a triangle topological structure, and then a simplest coarsened reduction map G' is obtained; by the adoption of a community finding algorithm based on genetics, initial community finding is conducted on the simplest coarsened reduction map G', and then a community finding result is generated; the community finding result is coarsened reversely layer by layer and restored to the map G, fine adjustment and optimizing processing is conducted accordingly, and then a community structure of the map G is acquired. According to the method and device, community finding and data analysis of the large-scale social networks can be accurately and efficiently achieved.

The invention provides a GPU-based Cache simulator and a spatial parallel acceleration method thereof. The Cache simulator comprises a Cache initialization module, a Trace preprocessing module, a Trace radix sorting module, a Cache simulation module and a Cache simulation result summarization and display module. The invention adopts powerful parallel processing resources of the GPU to preprocess memory access sequences, and utilizes the relevance of Cache configuration files to parallel sort the memory access sequences through a radix sorting algorithm operating on the GPU, thereby obtaining continuous memory access sequences mapped in each group and achieving inter-group parallel Cache simulation. Moreover, due to the parallel simulation algorithm with different levels and different grain sizes, the invention effectively reduces the Cache simulation time, improves the Cache simulation efficiency and realizes the accurate simulation of behaviors of the Cache. Moreover, the invention has the advantages of simple development environment and easy implementation.

Aspects for improving response time for database query execution are described. The aspects include comparing an estimate of sort pages needed for a database query against at least one threshold. A determination of whether to perform a parallel sort for the database query based on the comparison occurs, wherein response time improvement by the parallel sort is balanced against resource overhead of the parallel sort.

The invention relates to logistics sorting equipment, in particular to an automatic sorting machine of box-packed articles. The automatic sorting machine comprises a stand (1), an article storage tank (9), a sorting mechanism (10), a sliding baffle mechanism, a conveyor (7), an electric control cabinet (8) and a shell shield plate (2). The automatic sorting machine is characterized in that the article storage tank (9) includes two half tank bodies folded into the shape of a right angle, wherein the sections of the two half tank bodies are in mirror lateral symmetry, and the two half tank bodies are arrayed in parallel and form side rake angles with a horizontal plane; the high ends of the tank bodies form side rake angles beta with the horizontal plane, and the low ends of the same form side rake angles alpha with the horizontal plane; the low end of the article storage tank (9) is provided with the sorting mechanism (10); and the conveyor (7) is arranged below the sorting mechanism (10). The invention has high sorting efficiency because a sorting device adopts a completely parallel sorting mode and is especially suitable for multi-kind large-scale sorting occasions; in addition, the invention adopts an electromagnet as the power of the sorting mechanism (10) and has simple structure, low power and failure rate and reliable operation.

The invention discloses a low-latency butterfly assembly line sorting method based on an FPGA, belonging to the technology of radar signal processing, particularly relating to digital signal sorting technology. A butterfly processing unit divides reference unit sequences into groups according to the odd and even numbers of elements, 'in-group' and 'inter-group' sorting are carried out respectively, then a butterfly processor is subjected to assembly line reuse treatment, which means that data is subjected to parallel traverse sorting until the whole sorting is ended after the operation of n-1 hours. According to the method, through using the parallel sorting, the signal processing speed is greatly improved, the processing delay is reduced, and the disadvantage that too much time is consumed by traditional OS CFAR processing is overcome.

The invention provides an FPGA (Field Programmable Gate Array) parallel sorting method and system based on an index tree and a data linked list. The method comprises the following steps of reading a characteristic value and an identification value to be sorted and recorded from a memory in sequence; creating an identification value node in a data linked list storage area, and storing the recorded identification value in the identification value node; according to the recorded characteristic value, establishing a characteristic value node on the index tree of an index tree storage area; after all records are read and processed, traversing the characteristic value node of each layer on the index tree in sequence, and recovering the characteristic value according to a path of the characteristic value node; according to an identification value node address recorded in the obtained characteristic value node, finding the corresponding identification value in the data linked list storage area; finally, according to a traversing sequence, outputting all the recorded characteristic values and identification values. According to the FPGA parallel sorting method and system based on the index tree and the data linked list, data is coded/ decoded through the FPGA to realize data sorting so as to carry out data sorting while data is read.

The invention provides a method for performing parallel data processing in order. Parallel data processing is completed through N processing function modules, wherein N is a natural number greater than 1. The method comprises the steps that disordered records are received in M groups, and the records in each group are ordered, wherein M is a natural number greater than 1; the ordered records in each group are distributed to N queues corresponding to the group; each group corresponds to identical N queues; the N queues of each group correspond to the N processing function modules, and each processing function module corresponds to M queues; and the records in the M queues corresponding to each processing function module are output to the corresponding processing function module in order. Through the technical scheme, parallel ordering of the disordered records is realized, it is easy to increase processing speed by adding groups when ordering speed is low, the extensibility in the dataprocessing process is improved, and a single point of failure is avoided.

The invention discloses a SpaceFibre satellite-borne network service quality control system based on an FPGA (Field Programmable Gate Array), which comprises a multi-path virtual channel flow controlmodule and a single-path medium access controller, and is characterized in that the medium access controller is provided with a plurality of priority calculation modules and a scheduling module; a virtual channel flow control module which is used for receiving data of a virtual channel and generating a data sending request, and is also used for sending the data segment and the data word count value output by the buffer area to the media access controller when receiving a sending enable signal of the media access controller; a priority calculation module which is used for calculating the priority and sending the priority to the scheduling module; a scheduling module which is used for carrying out parallel sorting on the received priorities and selecting a scheduled virtual channel from allthe virtual channels according to a sorting result; transmitting and sending a data enabling signal to a virtual channel flow control module corresponding to the scheduled virtual channel; and outputting the number of the scheduled virtual channel and the data segment output by the buffer area of the corresponding virtual channel flow control module.

A data driven parallel sorting method includes distributing input data records to n partitions one by one in a circular manner. Each partition corresponds to a parallel sorting process with an allocated memory chunk sized to store m data records. The method also includes sorting, in parallel, current data records in respective memory chunks in respective partitions. The method also includes in response to distribution of data records of └m/n┘ rounds, circularly controlling one of the n partitions, and writing data records that have been sorted in the memory chunk of the partition into a mass storage as an ordered data chunk, and emptying the memory chunk. The method also includes in response to all data records being distributed, writing data chunks that have been sorted in respective memory chunks into the mass storage, and performing a merge sort on all ordered data chunks in the mass storage.

The invention discloses a four-DOF (degree of freedom) parallel sorting robot and belongs to the technical field of industrial robots. The robot comprises a rack, an end effector, a first branched chain, a second branched chain and a synchronous belt branched chain, wherein the first branched chain comprises a first driving arm, a first driven arm, a second driving arm and a connecting rod; the second branched chain comprises a third driving arm, a third driven arm and a third secondary driven arm; the synchronous belt branched chain comprises belt pulleys and synchronous belts; the first branched chain and the second branched chain are connected with driving devices through the driving arms respectively; the synchronous belt branched chain is connected with a fourth driving device through a first belt pulley; the first branched chain, the second branched chain and the synchronous belt branched chain are connected with the end effector. The robot has the characteristics of simplicity in manufacturing, convenience in mounting and debugging, lower manufacturing cost and the like.

The invention discloses a modular multi-level converter sub-module voltage sorting method. The method includes: using a programmable logic device to collect the voltage data of each sub-module, and setting N cascaded comparison and exchange units in the programmable logic device; allowing the collected voltage to serially pass N comparison and exchange units, saving the data in a tmp register when the data enters the comparison and exchange units for the first time, and comparing the data subsequently entering the comparison and exchange units with the data in the tmp register; outputting the current data to next-level comparison and exchange units after comparison or saving the current data in the tmp register to achieve voltage sorting. Compared with traditional sorting methods, the method has the advantages that the N cascaded comparison and exchange units are used for parallel sorting, the time used by sorting the voltage of N modules is 2N clock cycles, the method is easy to understand and implement, fast in calculation, and satisfies the requirements of voltage sorting of a modular multi-level converter with increasing levels.

The invention discloses a machine vision parallel sorting robot. The machine vision parallel sorting robot comprises a box body assembly (1). A robot horizontally-moving platform (7) is fixedly connected to the upper portion of the box body assembly (1). A driving connecting rod (9) is arranged parallel to the robot horizontally-moving platform (7). The driving connecting rod (9) is connected with connecting rods (10). Gaps between the connecting rods (10) are eliminated through a crossed steel wire (11). In the installation process, the gaps between the connecting rods are eliminated through pre-tightening of the crossed steel wire, abrasion is effectively reduced, the algorithm of the parallel robot is optimized, a front arm speed reducer and a servo motor are integrally designed, and the speed and accuracy of the robot are improved.

The invention provides a parallel acceleration method for KNN algorithm-based network intrusion detection. A CUDA parallel computing model is adopted in the method. The method comprises the steps thatfirstly, parallel analysis is conducted on KNN algorithm-based network intrusion detection, and when the distances between network intrusion detection data points and a training data set are computed, a general matrix multiplication function provided by the CUDA is adopted for acceleration, so that the computing speed is increased; secondly, in the distance sorting stage, a selection mechanism for two parallel sorting strategies is provided, the sorting algorithm needing the shorter sorting time can be flexibly selected for distance sorting according to the sorting result of a few data; and finally, in the classification stage of the intrusion detection data points, counting is conducted by adopting a CUDA-based atom addition operation. The experiment result shows that the acceleration method is valid, and the parallel acceleration performance of network intrusion detection is effectively improved on the condition that the detection rate is guaranteed.

The invention discloses a multi-channel multi-choice (MCMC)-based parallel sorting method. The sorting method comprises the following steps of: calculating likelihood estimation according to an initial state; calculating the posterior probability of parameters according to the likelihood estimation; performing MCMC simulation operation according to the posterior probability, and generating a novel state by taking the current state as the basis; calculating the acceptance probability according to the novel state, generating a first random number, taking the state at the next moment as a novel state when the first random number is less than the acceptance probability, otherwise keeping the current state invariable; generating a mark number of a Markov chain prepared to be exchanged in the same series of processors; and calculating the exchange probability when the Markov chain in the processor takes participates in exchange, generating a second random number, judging a comparison result of the exchange probability and the second random number, exchanging the heating parameters of the Markov chain during processing when the second random number is less than the exchange probability, otherwise avoiding exchange. According to the method, the execution time of an MC3 algorithm and an MCMC algorithm is shortened, and the communication overhead is reduced.

The invention discloses a commercial building energy consumption prediction optimization method and system, wherein the method comprises the steps: carrying out pre-screening of the influence factors of the commercial building energy consumption according to the importance degree, and obtaining the main influence indexes of the commercial building energy consumption; constructing an energy consumption sample set and dividing the energy consumption sample set into a training set and a test set; constructing a commercial building energy consumption random forest model, and determining parameters influencing the prediction precision of the random forest model; adopting a parallel ant colony algorithm, optimizing the parameters influencing the random forest model by using the data of the training set, obtaining the optimized parameters influencing the prediction precision of the random forest model, and thus obtaining the optimized commercial building random forest model; and substituting the data of the test set into the optimized commercial building random forest model, and carrying out energy consumption prediction to obtain a commercial building energy consumption prediction result. According to the method and system, the ant colony algorithm is improved by introducing the idea of parallel sorting, the problem of local convergence in the random forest multi-parameter optimization process is solved, and the convergence speed and the quality of the optimal solution are improved.

The embodiment of the invention discloses a symmetric matrix construction method for a compressed sparse matrix based on a GPU. The symmetric matrix construction method for the compressed sparse matrix based on the GPU comprises taking a compressed sparse matrix M in a given CSR-based storage format as an input matrix; according to the compressed sparse matrix M, converting the CSR storage format into a triple array T1 in parallel; storing each triple and a corresponding triple thereof in the triple array T1 in parallel, and carrying out parallel sorting to obtain a triple array T3; finding repeated data in the T3, constructing F-labeled repeatedly stored array elements, and deleting repeatedly stored elements with the F label of 1 in the T3 in parallel, so that a triple array T4 without repeated items is obtained; and according to the triple array T4, converting into the CSR storage format in parallel to be taken as an output matrix. By implementing the symmetric matrix construction method disclosed by the invention, processing performance of solving a symmetric matrix of the sparse matrix can be effectively improved, so that each step has parallelizability, and efficient parallel processing capacity in the GPU is played.

The invention relates to a four-freedom-degree parallel sorting robot driven by rotation pairs. The provided robot needs to have the beneficial effects of being compact in structure, convenient to install, high in responding speed and low in price. According to the technical scheme, the four-freedom-degree parallel sorting robot driven by the rotation pairs is characterized in that the robot comprises a tail end executer, a base and two branch mechanisms connected between the tail end executer and the base in parallel so as to drive the tail end executer to move; each branch mechanism comprises a spiral pair, a first connecting rod, the first rotation pair, a second connecting rod and two sub-branch mechanisms, all of which are sequentially connected between the tail end executer and the base, wherein the sub-branch mechanisms are connected in parallel; the spiral pairs in the two branch mechanisms are coaxial; and each sub-branch mechanism comprises the second rotation pairs, third connecting rods, the third rotation pairs, fourth connecting rods and the fourth rotation pairs, all of which are sequentially connected between the second connecting rods and the base.

The invention discloses a multi-column data sorting method, a device, a readable storage medium and electronic equipment. Different combination schemes are determined for different data tables, the data tables are divided into a plurality of columns to be sorted according to the selected combination schemes, execution expenditure in the sorting process is reduced, and meanwhile, by means of high concurrency of a graphic processor, the sorting efficiency of the data tables is improved. The data in each group are sorted in parallel after the data is grouped. According to the embodiment of the invention, the occupied memory is small, no large memory overhead is caused, and the execution efficiency is high.

The invention provides a parallel communication route establishment method and system, and the method comprises the steps that a transmitting end and a receiving end establish global indexes of all grid points for the same grid, and each process builds a grid point local-global mapping table based on the global indexes of the grid points; the sending end/receiving end performs parallel sorting oneach item in the grid point local-global mapping table among all processes of the sending end/receiving end; the processes of the sending end and the receiving end cooperate with each other, and a sending end-receiving end grid point mapping relation table is established in parallel; the sending end/receiving end performs parallel sorting on all the processes of the sending end/receiving end in the sending end-receiving end grid point mapping relation table; and each process of the sending end and the receiving end generates a communication routing relationship between each process and the corresponding receiving end or the process of the sending end according to the local sending end-receiving end grid point mapping relationship table of the processes. According to the method, the parallel communication routing relationship between numerical programs can be quickly established, and the method is efficient and has parallel expandability.

The invention discloses a parallel POI (Points of Interest) simplification oriented task splitting and distribution method which particularly comprises the following steps that (1), POI data is read; (2), the POI number Nf of target data is computed; (3), the number N of computation nodes of parallel computation is set; (4), a parallel POI simplification oriented data granularity model is established; (5) a Hi set (which represents a set of C (i, j) of a circle taking a point i as a circle center and C (i, j) as a radius internally tangent to a circle taking a point j as a circle center and C (i, j) as a radius) of each POI point is subjected to the parallel computation; (6), a relation between the number Nr of the residual points and the POI number Nf of the target data is judged, if Nr is greater than Nf, Step (7) is executed, and if Nr is less than Nf, the step is completed; (7), Hi of the POI point in each computation node is subjected to parallel sorting; (8), a minimum value Gc of the global Hi is subjected to parallel solving; and (9), the POI point corresponding to the minimum value Gc is selected, Step (6) is executed, and Step (6) to Step (9) are repeated till completion. The method has the characteristics of simplification, automation and intelligentization, can improve the efficiency of parallel POI simplification, and can guarantee the accuracy of the parallel POI simplification.