在分布式计算模式中，网格Grid和点对点P2P模型成为当今的主流技术，现有Grid和P2P的结合工作大多数都从节点互联、拓扑结构、DHT算法改进等较低网络层次或局部技术层面来进行，由于这两个模型在体系结构、工作模式、资源描述和查找机制上存在较多的异构性，因此二者的有机结合需要借助于本体技术实现语义层面的互联和互操作。本研究针对P2P-Grid应用环境的特点，探索在分布式场景下包含有大量动态不确定性的网络环境中，如何应用本体技术来解决P2P和Grid结合中节点标识、功能描述、资源查找等关键技术，改善现有P2P-Grid结合模式中各节点间知识和信息缺乏有效沟通基础和互理解手段的现状，重点研究本体模型的分布式工作和应用模式、动态不确定性的描述和处理机制、语义定量化度量和相似性推理等基础问题，力图为网络环境中各类分布式计算模式的结合提供互理解和互操作的基础理论和方法。

Nowadays, Grid and Peer-to-Peer (P2P) have played very important roles in distributed computing research field. In current research works, the combination of Grid and P2P almost investigates on the way of node interconnection, topology structure, distributed hash table algorithm improve, etc. Most of these integration methods are lower level connections or partial technical combination, because there are many heterogenities of Grid and P2P in aspects of architecture, work mode, resource description and retrieval. Therefore, the ideal combination of these two computation models should be the interconnection and interoperation based on ontology in semantic level. This research makes investigation into the solutions of node identification, function description and resource retrieval based on ontology, in allusion to the characteristics of distribution and dynamic uncertainty in P2P-Grid combination environment. It focuses on the basic issues of ontology distributed operation mode, dynamic uncertainty description and processing, quantitative measurement of semantic relations and fuzzy inference, and so on, in order to improve the current situation of knowledge and resource fragmentation in P2P-Grid without effective inter-comprehension means. These efforts try to provide the basic theory and methodology of inter-comprehension and inter-operation of different computational models in distributed computation environment.