58 results about "Distributed retrieval" patented technology

A workstation system produces a display presentation of a selected performance composition (e.g., a musical composition) responsive to composition data and responsive to one or both of input variables and a selected operating mode. The workstation can communicate with one or more external devices, such as other workstations, etc. The display system provides for selection of original compositions, creation of derivative compositions, distribution of compositions, monitoring of each performer's performance, group virtual performances, and for local and distributed retrieval and editing, which for music includes things such as changing keys, pitch, tempo, and other parameters. The musical composition's transformation can be performed locally or at the central or distributed music database. The musical composition data can be transposed via a controller, and can be transmitted to a plurality of the individual workstations that then display the musical composition. In one embodiment, a display system for use by a plurality of users provides a plurality of display presentations of a selected musical composition. The system is comprised of a plurality of individual workstations, each workstation comprising a communication interface providing for communications with the respective workstation of music data representative of the selected musical composition, memory for locally storing the data responsive to the communications interface, and a display apparatus provides a local visual display presentation representative of the selected musical composition responsive to the stored data. The system further provides for synchronizing the presentation on the plurality of local visual display presentations of the selected musical composition.

A musical workstation system produces a display presentation in one of a musical composition responsive to musical composition data and responsive to one or both of input variables and a selected operating mode. The system is comprised of (1) means to provide the musical composition data (such as local storage (ROM, RAM, CD-ROM, hard disk etc.), or via a communications interface to an external device (such as another music workstation, a master controller, a computer), a memory, a selection subsystem, a controller, and a display subsystem. The memory selectively stores the received original musical compositions. The selection subsystem determines a selected operating mode and display format. The controller, responsive to the selection subsystem, provides means for selectively controlling the storing of the musical composition data in memory and selectively processing (e.g. altering) the stored musical composition data responsive to the selected operating mode and the input variables to produce a particular one of a plurality of processed results, such as external communications, operating, mode, transformation to derivative musical compositions, etc. The music workstation can coordinate communications, or be a slave device, with one or more external devices, such as other music workstations, a master workstation, a controller, etc. The display system provides for selection of original compositions, creation of derivative compositions, distribution of compositions, monitoring of each performer's performance, group virtual performances, and also allows for local and distributed retrieval and editing of musical compositions.

A method is presented for transferring data objects between federated entities within a federation using artifacts. A first federated entity, such as a service provider, receives artifacts from a second federated entity, such as an identity provider, which generates data objects, such as assertions, for use at or by the first federated entity. The artifact references a data object that is locally stored by the second federated entity, which is implemented as a distributed data processing system with a set of data processing systems, each of which can generate artifacts and artifact-referenced data objects, and each of which can proxy data object retrieval requests to other data processing systems within the distributed data processing system. When the second federated entity receives a data object retrieval request with an artifact from the first federated entity, the artifact-referenced data object is retrieved from within the distributed data processing system using the artifact.

The invention discloses an intelligent urban construction examining and approving method based on a case-based reasoning technology. The intelligent urban construction examining and approving method based on the case-based reasoning technology comprises the following steps of constructing an examining and approving case library; inputting new examining and approving case and model parameter information; submitting jogs to a Hadoop cluster to search KNN (k-nearest neighbor algorithm) Map Reduce cases; statistically analyzing a searching result on the basis of a 'weighted integral model'; evaluating and correcting the cases; and performing distributed full-text searching on examining and approving data. The intelligent urban construction examining and approving method has the advantages that by the method, the circumstance of manual examination and approval application at present can be changed, the work efficiency is improved, the basis on examining and approving is increased, and an examining and approving process is intelligent. Distributed searching can be carried out by using a Hadoop frame and a MapReduce frame through a cloud computing center, and a distributed case searching model based on the case-based reasoning technology is established. The 'weighted integral model' is creatively raised to statistically analyzing searched similar cases, and a guidance which is beneficial to new examining and approving cases is obtained.

A resource manager allocates operation requests to security network devices according to device characteristics. The devices collect and/or manage data from an environment. The devices may include a camera, a multimedia recorder, an analyzer, and a meta-data server. One or more users submit operation requests to a controller. The controller allocates the network devices to the operation requests according to the device characteristics. The device characteristics include availability of the device, media flow data of the device, location of the device, and capabilities of the device.

The invention discloses a POS (Point Of Sale) system based on a cloud middleware. The POS system comprises a cloud POS terminal, an encrypted tunnel, a financial cloud middleware and a cloud server, wherein the cloud POS terminal is provided with an embedded safety operation system for configuring user account information, and inputting and displaying user consumption information; the cloud POS terminal is provided with a third-party application; the encrypted tunnel is used for encrypting the data; the financial cloud middleware is used for implementing data exchange between the cloud server and the cloud POS terminal through protocol conversion and standard conversion; the cloud server is provided with a POS transaction big data real-time processing platform for carrying out data acquisition sharing, data integration, distributed retrieval, load balancing and data management on the data received by the cloud server; a data processing result of the POS transaction big data real-time processing platform is transmitted to the cloud POS through the third-party application. According to the POS system, function extension of the POS terminal is improved to provide more complete functions for financial service.

The invention provides a deep hash-based medical image distributed retrieval method. The method comprises deep hash extraction image features and Hadoop-based batch image feature matching parallel computing; in the deep hash extraction image features, similar or dissimilar image pairs are used as training input through a convolutional neural network model, the gradient of an objective function relative to multilayer network weights is calculated by use of a back propagation algorithm, and finally multiple output values of each image are guided to be approximate to discrete 0 or 1; and in the Hadoop-based batch image feature matching parallel computing, a feature file of batch images is divided into a plurality of blocks, the blocks are independent from each other, the blocks are distributed to different nodes for execution through an Apache Hadoop YARN (Yet Another Resource Negotiator) resource manager, and finally results executed by all Mappers are combined into a Reducer. According to the method provided by the invention, the gap between image representation and semanteme can be reduced, the retrieval precision is improved, the retrieval process is accelerated through parallel feature matching, and the retrieval efficiency of batch medical images is enhanced.

The invention discloses a directory distribution method and network architecture orienting to a high-concurrency retrieval system, and belongs to the field of computer network. The invention provides a retrieval method combining the integrated retrieval based on servers with the distributed retrieval based on directory distribution according to the behavior characteristics of video retrieval of users, and makes full use of the server capability of each user node to complete the update of the consistency in the directory file when the user node resource is free, so that the user can acquire data consistent with those on the retrieval server when retrieving. The user firstly searches for the required retrieval result by reading the local directory file when retrieving, and adopts the integrated retrieval based on the retrieval server when the directory file cannot meet the requirements of the user, thereby not only guaranteeing the retrieval response time and improving the user experience, but also guaranteeing the accuracy and integrity of the retrieval result. Under the pressure of high-load and high-concurrency retrieval access, the invention, which has high expandability, can obviously relieve the pressure of servers in a retrieval center, reduce the hardware input and reduce the system cost.

The invention provides a Hadoop-based distributed retrieval method for power transmission line monitoring videos. The method comprises the steps of based on a Hadoop platform, dividing massive video data storage into different video clips and storing the video clips in Blocks of Hadoop; extracting key frames under the assistance of OpenCV; performing similarity comparison on the key frames and a sample picture submitted by a user to obtain N key frames most similar to the sample picture, and obtaining the video clips corresponding to the key frames; and displaying a retrieval result in a Web interface, thereby finishing video data retrieval work. The distributed retrieval method provided by the invention has the advantages of quickness, high efficiency and expansibility in processing large-capacity data, and solves the high-performance demand problem of power transmission line monitoring video retrieval of large data volume.

The invention discloses a power distribution/utilization data retrieval method and system. The method comprises: carrying out dividing of a plurality of data storage nodes in a distributed database ora file system, and storing power distribution/utilization data in the data storage nodes corresponding thereto; establishing and managing index fragmentation files, which correspond to the data storage nodes and the power distribution/utilization data, by an index cluster system through a distributed index establishment mode; and retrieving power distribution/utilization data stored in a data storage node by a query cluster system on the basis of the index fragmentation files through a distributed retrieval mode, and providing a retrieval result. The power distribution/utilization data retrieval method and system of the invention can solve the index problem of massive data, have the characteristics of high throughput, high expansibility, high concurrency, high fault tolerance and the like, are suitable for concurrent accessing to a super large data set, reduce loads of a retrieval main-node, and can improve availability of query services, increase response speed of services, and improve query experience of a user.

The invention relates to a data retrieving and ordering method for an electrical power system. The data retrieving and ordering method comprises the following steps: (1) storing data by virtue of a distributed file system; (2) establishing a professional knowledge relevant synonym participle lexicon; (3) establishing full-text index for data; (4) carrying out retrieving and ordering in a parallelization manner; and (5) realizing the fuzzy matching of multiple retrieved fields. According to the method, the storing, retrieving and ordering of the data are realized in a parallel computing manner, so that the rapid retrieving is realized; by defining a professional word segmentation system, the indexing efficiency is improved; by virtue of a full-text retrieval manner and a distributed retrieval manner, the retrieval speed of a non-standard text is increased; by retrieving self-learning and synonym definition of a keyword, the fuzzy retrieval is realized; and the synonym lexicon is utilized for intelligently identifying situations of abbreviation of terms, nonstandard filling and the like, so that the business Integrity of a retrieval result is guaranteed. The method is simple and practical, the data of a relation database can be directly imported, and the distributed storage and the data index can be automatically established, so that rapid deployment and multi-scene application are realized.

The invention discloses a distributed retrieval resource library selection method based on a variational auto-encoder, and the method comprises the steps: building an encoder and decoder network structure by utilizing a deep neural network, learning implicit representation of a resource library text, and capturing deep semantic representation of the resource library text; reasoning the extended text of the query word through a model obtained by an unsupervised training method to obtain a hidden representation of the query word; obtaining the correlation ranking of the resource library by calculating the similarity between the query words and the implicit representation of the resource library. The model is unsupervised training, a resource library and a hidden representation vector of a text are automatically obtained, and the defect of text feature design in a supervised training method can be overcome. In addition, the network structure of the variational auto-encoder is simple, andthe calculation time consumption of variational reasoning is lower than that of an LDA topic model based on a Markov chain Monte Carlo reasoning method. And after model training is completed, the timeconsumption for resource library selection is low, and the resource library selection efficiency is high.

The invention discloses a distributed search method and device. The method comprises the steps that original data is classified according to a data subject, and each class of the original data is stored into a plurality of storage modules; the storage modules are subjected to distributed storage according to a first default rule; each storage module is searched according to default key words, a corresponding search result of each storage module is obtained, and the corresponding search results of all storage modules are merged into a final search result. The device comprises a data classification module, a distributed storage module and a search module. According to the distributed search method and device, the original data is classified, distributed storage and search are conducted, the search time is short, the hit rate is high, the search efficiency is improved without longitudinal extension and horizontal extension, and the software and hardware resource consumption is greatly reduced.

The invention discloses a distributed retrieval system and method oriented to industry metadata registration. The system comprises the following three modules: a registration module, a storage module, and a distributed retrieval module, wherein the registration module is used for realizing industry metadata registration and sharing across data centers, a data management unit of the data center is used for realizing the metadata self-audit and registration aiming at different industry central and local data centers, and the unit classifies and processes the managed data, injects partial metadata for retrieval into a central node, and all metadata is saved at the local; the data management unit generates inverted indexes for the managed metadata registration, the inverted indexes are distributed in the central and local nodes, and the separation of the data storage and the data management is realized at the same time; the distributed retrieval module is used for realizing the industry metadata sharing across the data centers, and performing distributed retrieval on mass of industry metadata in different data centers. Through the adoption of the system and method disclosed by the invention, the problem of industry metadata sharing across the data centers is solved, mass of industry metadata can be processed in a distributed manner, and the retrieval demand of a user is satisfied.

The invention discloses a distributed retrieval method and device. The method comprises the following steps that: after a retrieval request is received, according to the fingerprint of to-be-retrieved information carried in the retrieval request, searching a similar fingerprint in stored fingerprints; for each found similar fingerprint, independently carrying out the following operations that: correspondingly comparing the segment of the similar fingerprint and the segment of the fingerprint of the to-be-retrieved information in sequence according to a preset sequence, and stopping comparison after the first identical segment is found; comparing the identification of the identical segmentation with the identification carried in the retrieval request, and if the identification of the identical segmentation and the identification carried in the retrieval request are the same, ensuring that a counting result contains the similar fingerprint, wherein the division of the segment and the identification of each segment are determined according to a first rule; and returning the counting result. By use of the method, the problems that counting is unstable and additional counting duplicate removal operation is required when similar information identification is carried out in the distributed system can be solved.

The invention belongs to the technical field of real-time capture, storage and retrieval of network traffic, and particularly relates to a distributed network traffic retrieval method and device. The method comprises: acquiring data; recombining a session and extracting metadata; writing the session into a disk; and performing a distributed retrieval algorithm DSearch and data storage. The invention provides a distributed index establishment algorithm DIndex and a distributed retrieval algorithm DSearch, which can support distributed real-time high-speed network full-traffic storage, indexing and retrieval services, and satisfy the use requirements of users for high bandwidth and multiple branches.

The invention provides an unconscious intelligent retrieval method based on remote sensing space big data, which comprises the following steps: 1, firstly, performing storage, organization and sharing management on a remote sensing image data entity through an MFS distributed file system, and performing distributed storage on metadata and a thumb graph of a remote sensing image through a PostgreSQL to ensure the security of the data; 2, quickly importing the metadata in the PostgreSQL into an Elasticsearch cluster, creating an index, and building a distributed retrieval service for the remote sensing image metadata through an ES; 3, performing, by a user, space retrieval on the remote sensing image based on a three-dimensional digital earth or a two-dimensional GIS system, providing unconscious intelligent query by the system, specifically, utilizing an Elasticsearch distributed retrieval cluster for carrying out rapid retrieval on the metadata number, such that at the moment, the user does not need to carry out a large amount of operation interaction, and the inquired remote sensing image information result is calculated and returned in real time based on coordinate information of mouse movement or simply drawn retrieval areas.

The invention provides a data storage method and device based on big data, a storage medium and a processor. The method comprises the steps: establishing a basic storage database through a distributedcolumn storage database and a distributed retrieval engine, wherein the distributed column storage database is used for storing original data, and the distributed retrieval engine is used for storingindex information of the original data; constructing a data table according to a data storage structure and a data storage format in the basic storage database; dynamically managing the data table ina configuration library mode; and performing data storage based on the data table. According to the method, a basic storage database is established by adopting a distributed column storage database and a distributed retrieval engine, the rapid retrieval of data and large data throughput are achieved, a data table is constructed by using a standard data storage structure and a standard data storage format, then the dynamic management data table is managed in a configuration library mode, and the rapid storage of the data is achieved.