适用于高性能计算平台的金融风险度量模型及计算方法研究

The core foundation of financial risk management is the risk measurement. With the increasing instability in the financial markets, risk measure to deal with huge amounts of data, it is difficult to meet real-time requirements, this paper will solve this issue by technical research on calculable risk modeling and calculation method: (1) Model simulation and parallel processing of financial risk measurement; (2) It contains the core node, management node on the optimal architecture and research within the node four parallel computing architecture calculations; (3) Allocation and coordination mechanisms and task scheduling algorithm between nodes under massively parallel computing environments; (4) High-performance random number generator. By studying this subject, the use of high-performance computing system to achieve real-time measurement of financial risks, and promote the development of financial sector risk measurement techniques, to provide a theoretical basis for high-performance computing applications, and expand the financial sector massively parallel computing software application.

金融风险管理的核心基础是风险度量，随着金融市场不稳定性逐渐增加，风险度量需要处理海量数据，难以满足实时需求，本课题将开展创新的可计算风险建模与计算方法研究解决这一技术问题：（1）金融风险度量模型及新型并行化计算方法研究；（2）基于动态扩展与自适应网络环境的四层并行架构研究；（3）大规模并行计算环境下节点间任务动态分配与协调调度机制研究；（4）基于众核的高性能随机数产生器研究。通过本课题的研究，利用高性能计算平台实现金融风险实时度量，促进了金融领域风险度量技术的发展，为其在高性能计算应用提供理论依据，并拓展金融领域大规模并行计算软件应用。

金融风险管理的核心基础是风险度量，随着金融市场不稳定性逐渐增加，风险度量需要处理海量数据，难以满足实时需求，本课题将开展创新的可计算风险建模与计算方法研究解决这一技术问题：（1）金融风险度量模型及新型并行化计算方法研究；（2）基于动态扩展与自适应网络环境的四层并行架构研究；（3）大规模并行计算环境下节点间任务动态分配与协调调度机制研究；（4）基于众核的高性能随机数产生器研究。通过本课题的研究，利用高性能计算平台实现金融风险实时度量，促进了金融领域风险度量技术的发展，为其在高性能计算应用提供理论依据，并拓展金融领域大规模并行计算软件应用。