7results about How to "Reduce memory pressure" patented technology

The application provides an optimization method and system for data processing. The method comprises: obtaining a table association task to be executed; determining a current association mode of the table association task, and executing the table association task belonging to a current computing batch according to the current association mode; collecting intermediate result data generated by the current computing batch; analyzing the intermediate result data to determine a target association mode of the table association task in a next computing batch; and executing the next computing batch according to the target association mode. The application can adapt to changes in data size and improve the efficiency of big data table association calculation.

This application relates to a data storage method, apparatus, vehicle, electronic device, and storage medium. The method includes: determining the computation order of N operators included in a neural network; determining M operators from the N operators based on the computation order; calculating the i-th operator among the N operators according to the computation order to obtain the output of the i-th operator; if the i-th operator is one of the M operators, then saving the output of the i-th operator to a first type of storage space, wherein the output of the i-th operator saved in the first type of storage space cannot be deleted after calculating the (i+1)-th operator but before calculating the k-th operator; if the i-th operator is not one of the M operators, then saving the output of the i-th operator to a second type of storage space, wherein the output of the i-th operator saved in the second type of storage space can be deleted after calculating the (i+1)-th operator. This method can solve the problem of excessive memory consumption during the computation of a neural network model.

The embodiments of the present application provide a method, device, equipment and computer-readable storage medium for dynamic intelligent retrieval of an address book organization permission tree based on a compression algorithm. The method includes obtaining a string to be retrieved; based on the string, traversing the stored visible organization structure tree to obtain an initial retrieval result; calculating a similarity score based on the string and the initial retrieval result; and obtaining a target retrieval result based on the similarity score and the hotspot usage data of the current user that has been obtained. In this way, the server memory pressure can be effectively reduced, and the efficiency of the address book organization structure search and the intelligent retrieval capability can be improved.

The present invention provides a data storage monitoring method and system for a hydropower fusion terminal, and relates to the technical field of agricultural irrigation equipment. After detecting a data storage instruction, the present invention performs security and integrity detection on the data content of the data to be stored to determine whether the data to be stored is normal, and then determines the transmission time and transmission mode according to the storage priority, splits the data to be stored into multiple data packets, and sends them to the hydropower system platform. The present invention performs a security and integrity detection before data storage and transmission, so that the hydropower system platform can directly store data, reduces the memory pressure of the hydropower system platform, and improves data storage efficiency. It solves the problem of low efficiency of storing agricultural irrigation data on the hydropower system platform, and improves the data storage efficiency of agricultural irrigation data.

A neural network inference acceleration method and device are provided. The method includes: acquiring, by a memory having a computing unit, a video and a prompt word; acquiring, by the memory having the computing unit, at least one image related to the prompt word from the video; dividing, by the memory having the computing unit, each of the at least one image into a plurality of blocks; storing, by the memory having the computing unit, the plurality of blocks; storing, by the memory having the computing unit, position information of the plurality of blocks in each of the at least one image; acquiring, by a processor, each of the at least one image as the plurality of blocks and the position information; dividing, by the processor, each of the at least one image into a plurality of sub-blocks; acquiring, by the processor, from the memory having the computing unit, an embedded vector of a pre-stored sub-block corresponding to each of the plurality of sub-blocks, respectively; and performing an inference operation using the embedded vector and a first neural network.

This application provides an application installation method, apparatus, device, and medium based on compilation optimization, relating to the field of computer technology. The method includes: in response to an installation operation on a target application, obtaining an application package corresponding to the target application; extracting baseline configuration files for various scenarios included in the application package; installing the target application based on the application package, and compiling hotspot code corresponding to a first configuration file that meets the compilation requirements during installation from the baseline configuration files for each scenario; and, when it is determined that the electronic device is in an idle state, compiling hotspot code corresponding to a second configuration file (excluding the first configuration file) from the baseline configuration files for each scenario, thereby completing the installation of the target application. This method aims to save resources, improve performance, and enhance the user experience.

The present invention discloses a water surface rendering method, device, storage medium and computer equipment, which are mainly capable of improving the rendering efficiency of large-area water surfaces and ensuring the rendering effect of large-area water surfaces. The method includes: determining the inter-frame position offset of a virtual character located in a target virtual water surface in the world space, wherein the water surface model of the target virtual water surface includes a ring-shaped distant water surface model and a near water surface model nested in the distant water surface model and moving with the virtual character; according to the inter-frame position offset of the virtual character in the world space, determining the sampling offset when the near water surface model is perturbed, and determining the reverse offset of the vertices of the distant water surface model when the virtual character has a position offset; based on the sampling offset and the reverse offset, rendering the near water surface model and the distant water surface model respectively. The present invention is suitable for rendering large-area water surfaces.