Towards A Theory of Hierarchical Modeling

目前的性能和可靠性建模技术和工具本身被认为在很大程度上不足以获得对超大规模电子系统的行为的现实预测。许多作者在处理大型模型方面遇到了困难，并求助于各种专门的、特别的技术。该项目的重点是为大规模性能和可靠性模型的规范、组合、分解和解决方案开发高级形式化(语言)。期望这种模型合成的形式将有助于模型者组装这些大系统的模型，并有助于理解由替代的模型合成结构实现的误差和效率的提高。预计将出现一种自动模型管道安装系统。

Current performance and reliability modeling techniques and tools are seen to be, by themselves, largely inadequate for obtaining realistic predictions of the behavior of very large scale electronics systems. Numerous authors have experienced difficulties with tractability of large scale models, and have resorted to a variety of specialized, ad hoc techniques. The focus of this project is the development of a high-level formalism (language) for the specification, composition, decomposition, and solution of large scale performance and reliability models. It is expected that this formalism for model composition will aid the modeler in assembling models of these large scale systems and in understanding the errors as well as the gain in efficiency achieved by alternative model composition structures. An automated model pipefitting system is expected to emerge.