DECF: A Coarse-Grained Data-Parallel Programming Framework for Seismic Processing

摘要

In this paper, we present a Distributed Execution Control Framework (DECF) for seismic processing. The DECF framework employs a coarse-grained data-parallel execution model within a cluster environment and achieves high throughput. The DECF schedules processing modules written as sequential programs to run simultaneously on multiple computers, or multiple CPU cores. The DECF hides the messy details of scheduling the use of computers and their CPUs, recovering from communication or computer failures, and transporting data between processing modules.We also introduce a parallel implementation of 3D Pre-stack Kirchhoff Time Migration algorithm using MPI and share the practical lessons learned from the process of designing, implementing, and maintaining it, which are of great help to the future work of the DECF.