Multicore Mode The Seismic Prospecting Parallel Computing Research

Currently, many engineering computing needs handle large-scale mass data. How to handle such data quickly and efficiently has become utmost concern among technical personnel. Parallel processing technology can process large-scale mass data fast. In the field of oil exploration, With the current oil and gas exploration has been expanding, making exploration targets geological bodies in the surface more and more complex; same reservoir exploration and development efforts are deepened, it cause seismic exploration data quantity and Computation present exponential growth. All of this needs parallel processing technology urgently.In parallel processing technology, Message passing programming model (MPI) is the main parallel model. This model is influenced by two factors; it is network speed and single node computational performance. With the popularity of the multi-core processor, it provides Mining Spatial of CPU computing resources under Single node. The main purpose of article is to improve parallel efficiency by enhancing single node computational performance. In order to enhance the computing power of a single node, the article introduces OpenMP multi-threaded parallel programming. The OpenMP Application programming Interface is a defacto standard for parallel programming on shared memory multiprocessors. In the parallel technology, with the popularity of multi-core processors, OpenMP parallel programming will be used widely.Kriging is an important method in the seismic data processing.In this paper, according to crucial parts of kiriging, numerical experiments have been conducted under MPI,OpenMP and MPI+OpenMP conditions, which results shows that MPI+OpenMP is an effective Parallel Strategy, and possesses good expandable capability. Finally, we apply MPI+OpenMP to Kriging method, the experimental showed that:the parallel solution speed-up ratio reach 1.72.