非连续变形分析（DDA）方法在模拟岩体的大位移和破碎过程时具有天然的优势。然而该方法的瓶颈在于求解问题的规模不能太大、计算效率也不高。本项目基于成熟的商业云计算平台，研究并行的三维球颗粒DDA方法及其云计算模式。拟采用理论分析、算法推演和编程、试验验证等手段，研究如下内容：能够模拟岩体三维破裂过程及耦合应力、温度、渗流效应的三维球颗粒DDA算法；三维球颗粒DDA方法中的高效接触搜索算法、接触收敛准则和接触模型；三维球颗粒DDA方法的并行计算技术，包括稀疏矩阵存储方式和方程组并行迭代求解技术；云环境中并行执行的计算框架和动态并行作业调度方法；开发基于云计算平台的三维球颗粒DDA并行程序。采用多种手段对程序进行调试和验证。最终实现对大型岩体工程的破坏失稳过程的快速精确模拟，并推动云计算在岩体工程中的应用。

The discontinuous deformation analysis (DDA) method has inherent advantages when it is employed to simulate the large displacement and fracturing process of rock masses. The bottleneck of this method is that the computational scale is small, and the efficiency is low. This study is based on the mature commercial cloud computing platform, and develop the parallel spherical discontinuous deformation analysis and its cloud computing model. By applying theoretical analysis, algorithm formulation, programming and test verification, this project supposes to perform study on the following aspects: the 3D spherical DDA algorithm which can be used to simulate the fracturing process and coupling effect of mechanical, temperature, and seepage of rock masses; effective contact search algorithm, criterion of contact convergence, contact model; parallel computation technique, including the storage method of sparse matrix and parallel iterative solution method of equations; parallel framework and dynamic scheduling method in cloud; developing the parallel 3D spherical DDA program in cloud computing platform. Verification and validation will be done by means of classic benchmarking problems, laboratory tests and practical engineering problems. The objective of this research is to achieve the simulation analysis of failure process of large-scale rock engineering. Besides, this research will help to promote application of the cloud computing in rock engineering.