Abstract: One or more example embodiments are directed to a storing method of an embedding table, an apparatus, an electronic device, a non-transitory storage medium, and/or a system, the storing method including: acquiring an embedding table; and storing the embedding table in a memory of at least one CXL-PNM device, the storing enabling the at least one CXL-PNM device to perform parallel lookup of the embedding table in the memory using processing circuitry of the at least one CXL-PNM device, wherein the at least one CXL-PNM device includes the memory and the processing circuitry.

Abstract: A computing method for a convolutional neural network performed by a first device may include executing a first convolution layer of the convolutional neural network to obtain a first convolution matrix, writing the first convolution matrix to a high bandwidth memory (HBM) of a second device, controlling the second device to perform an activation operation on the first convolution matrix in the HBM through a first activation layer following the first convolution layer to obtain a first activation matrix, in response to the first activation matrix being written to the HBM, controlling the second device to perform a pooling operation on the first activation matrix in the HBM through a first pooling layer following the first activation layer to obtain a first pooling matrix, and in response to the first pooling matrix being written to the HBM, executing a second convolution layer following the first pooling layer

Abstract: A method and device for data storage are provided. The method includes determining, based on a write message corresponding to a first file, an application corresponding to the first file, a file type of the first file, and first attribute information corresponding to the file type, the first attribute information corresponding to the file type obtained based on second attribute information corresponding to all files of the file type on which the write operation processing have been performed in response to the respective write messages of the application, predicting a stream identification information corresponding to the first file by using a file classifier based on the application corresponding to the first file, the file type of the first file and the first attribute information corresponding to the file type, and sending a writing command carrying the stream identification information to a storage device.

Abstract: A method and device for data storage are provided. The method includes determining, based on a write message corresponding to a first file, an application corresponding to the first file, a file type of the first file, and first attribute information corresponding to the file type, the first attribute information corresponding to the file type obtained based on second attribute information corresponding to all files of the file type on which the write operation processing have been performed in response to the respective write messages of the application, predicting a stream identification information corresponding to the first file by using a file classifier based on the application corresponding to the first file, the file type of the first file and the first attribute information corresponding to the file type, and sending a writing command carrying the stream identification information to a storage device.