A Distributed Architecture for Global WWW Cache Integration

该奖项支持一个为期三年的项目，以继续调查互联网网页缓存。 这是一个为期两年的成功项目的后续项目，该项目在五个NSF资助的超级计算中心和FIX-West部署了缓存服务器。 该项目得到了相当多的国际参与，并产生了名为Squid的早期缓存软件，目前已在400多个站点使用。 这个后续项目做了以下工作： ...鼓励美国更多的参与。 ...继续Squid代码开发，并维护公共领域的代码库以进行实验和协作。 ...至少举办一次研讨会，以促进缓存研究人员之间的思想交流。 ...标准化互联网缓存协议（ICP），以确保实现之间的互操作性。 ...开发自动缓存发现系统。 ...协助改进网络软件http，以支持缓存。 ...与其他缓存研究小组合作，使用多播在缓存之间分发数据。 ...制定高性能缓存硬件的规格，并与供应商合作开发此类设备。 ...调查研究问题，如缩放、记录系统性能和信息提供程序命中统计数据，以及改进统计数据收集以实现更好的缓存模拟。

This award supports a three year project to continue investigation of Internet web caching. It is a follow-on to a successful two year project which deployed caching servers at the five NSF-funded supercomputing centers and at FIX-West. The project resulted in considerable international participation and an early version of caching software called Squid which is now being used at over 400 sites. This follow-on project does the following: ... encourage greater U.S. participation. ... continue Squid code development and maintain the public domain code base for experimentation and collaboration. ... hold at least one workshop to improve exchange of ideas among caching researchers. ... standardize the Internet Caching Protocol (ICP) to assure interoperability between implementations. ... develop an automatic cache discovery system. ... assist in improving the web software, http, to support caching. ... collaborate with other caching research groups to use multicast for distributing data between caches. ... develop specifications for high performance caching hardware, and work with the vendors in developing such equipment. ... investigate research questions such as scaling, recording system performance and information provider hit statistics, and improving statistics gathering to enable better caching simulations.