Two-layer dynamic scheduling method facing to ensemble prediction applications

Abstract

The invention discloses a two-layer dynamic scheduling method facing to ensemble prediction applications, aiming at providing a two-layer scheduling method based on dynamic selection among mesh nodes and the dynamic designation of the resource number in the node. The technical scheme is as follows: two-layer dynamic scheduling system is established firstly; furthermore, the system is arranged at the server terminal and each mesh node terminal; the two-layer scheduling system initializes the scheduling process of the current ensemble prediction; the mode prediction service at each mesh node dynamically competes the unconsumed initial sample data file to the sample data management service; the mesh nodes optimize the resource number of the appointed mode prediction program and start the mode prediction program; the sample data management service at the server terminal stops the ensemble prediction scheduling process and starts the postprocessing on the ensemble prediction flow. The two-layer dynamic scheduling method has real-time dynamic characteristic. By adopting the two-layer dynamic scheduling method, the time effectiveness of the ensemble prediction with large-scale calculation characteristic can be improved, and the calculation cost for executing the ensemble prediction once can be saved effectively.

Description

technical field [0001] The invention relates to a dynamic scheduling method based on grid technology, in particular to a dynamic scheduling method oriented to the application field of ensemble forecasting. Background technique [0002] Numerical weather prediction is an indispensable method in the operation of meteorological operations. It is based on the physical laws of atmospheric motion under the given initial conditions and boundary conditions, and numerically solves the basic equations of atmospheric motion to predict the atmospheric state process in the future. . The general numerical weather prediction process includes observational data preprocessing, meteorological data variational assimilation, model forecasting, post-processing, product visualization, etc. These tasks, especially model forecasting tasks, require a large number of high-performance resources for numerical model calculations. Ensemble prediction is a new numerical weather prediction technology prop...

AI Technical Summary

Problems solved by technology

[0007] 4. The parallelism and speedup ratio of the model forecasting program after the initial perturbation do not increase linearly with the number of CPU resources N (N is generally a power of 2)

On the other hand, these scheduling methods do not take into account the dependence of the parallel application itself on the number of CPU resources, such as parallelism, speedup, etc. These methods are difficult to solve the special application problem of ensemble forecasting with large-scale calculations, especially It is this kind of application problem that the computing power of computing node resources changes dynamically due to the long computing time.

Images