The web became an important platform in the past years. Not long ago, it was used only for publishing simple documents, and now it's the primary platform of content distribution, social communication and shopping for many. The content, the standards that define the web and the browsers implementing them were developed hand in hand. There is a never ending race between browser companies to achieve better performance and usability, and that leads to a constantly improving overall user experience. This is amplified by the improving speed of internet connections.
But browsers are not the only player in this game. Web authors, hosting companies and internet service providers all play special role in improving the performance and user experience on the web. All of them have a different view of the same process, and different means to influence and improve it.
One of the methods that intermediaries can use is prefetching web content to a local cache to make it instantly accessible when it is needed. It is also possible to push the content from the proxy to the client if needed. The topic of this thesis is the examination of existing optimizations of this type, and exploration and implementation of new type of optimizations based on it. I will answer the questions concerning the limitations of content prefetching, and look at the advantages, disadvantages and performance characteristics of the means of it. The new prefetch methods will be tested and verified with a prototype implementation using a simulated network.