Abstract: The disclosure relates to a method and a system for predicting the operation time of sparse matrix vector multiplication. The method comprises constructing a convolutional neural network comprising an input layer, a feature processing layer, a data splicing layer and an output layer for outputting prediction results. The method further comprises acquiring a plurality of groups of sparse matrices with known sparse matrix vector multiplication operation time as sample data, inputting the sample data into the convolutional neural network to train the convolutional neural network, and inputting the sparse matrix to be classified into the trained convolutional neural network to realize the prediction of the operation time of sparse matrix vector multiplication.

Abstract: A method and device for simulating microstructure evolution of a material based on solution in an exponential time-difference format. The method includes: establishing a reaction rate theory model for substance defects, wherein the model is expressed with equations that comprise linear terms having coefficients characterized with matrixes; and iteratively solving the equations by using an exponential time-difference format, wherein during the iterative solving, the linear terms with exponential powers of the matrixes as the coefficients are integrated. Since a rate theory is not limited by spatial-temporal scales, the advantages of the rate theory can be significantly reflected when the microstructure evolution is simulated under a high damage dose condition; and then, the equations are solved by using the exponential time-difference format, with a solved result better in accuracy and higher in precision.