On Tuesday 01 July 2003 15.52, Bhattacharyya, Somraj wrote: > Hi guys !! > > "If we replicate a web server and place it near as near as possible > to a client then we might not require caching servers." This is a > general statement and vissible for very large and popular web > servers.
Yes and no. This is the basic idea of Content Distribution Networks. Different Content Distribution Networks uses different methods of distributing the content, but most runs as a hybrid of pushing the content and caching. You can easily push the content to a Squid surrogate server by requesting the content from it. > Please note: Say by using say 5000 caching servers (roughly)we are > serving a country, where same web pages are cached in dfferent > severs.May be two servers side by side on a same table containing > the same page, solving problem but isnt it wasting resources.But > one replicated server of a particular website could handle the same > situation, but ofcourse only for that web site. There is many approaches to this issue. The two fundamental approaches is that either you make sure the clients only hit one of the servers, or you make the servers soemwhat aware what the other server has in it's cache. If using Squid then ICP or cache-digests with proxy-only peerings come to mind. > How can squid and apache can technically help me to support this > idea.Mainly to support replicating web doucuments with in web > servers not caching. The difference between replication and caching is merely a technical difference, the functionality is mostly the same. Start by defining the function you want to provide, then select technology to use. -- Donations welcome if you consider my Free Squid support helpful. https://www.paypal.com/xclick/business=hno%40squid-cache.org If you need commercial Squid support or cost effective Squid or firewall appliances please refer to MARA Systems AB, Sweden http://www.marasystems.com/, [EMAIL PROTECTED]
