分布式虚拟环境AIMNET的关键技术概述

AIMNET:The Key Techniques in Distributed Virtual Environment

本文概要介绍南京大学自主研究开发的分布式虚拟环境AI MNET的关键技术。AI MNET借鉴了兴趣管理的思想,提出了双向-共享组播树的通讯结构,从而大大降低了网络通讯量,并保证了AI MNET系统良好的扩展性;AI MNET中静态数据采用分布式存储结构,同时为了实现场景的远程渲染而采用了负载均衡、数据压缩、累进式渲染、航位预测算法以及数据预取等技术。

This paper introduces the key techniques in distributed virtual environment AIMNET developed by Nanjing University. Using the methods of Interest Management by reference, AIMNET advances the communication topology structure of bi-directional shared multicast tree, in order to reduce the number of messages and emphasize the scalability of the system. At the same time, the distribution of static data is presented in this paper, in company with the remote real-time rendering technology, which includes load balancing, data-compression, progress rendering, dead reckoning and data pre-fetching.