Is there any request prediction and preload solutions for web app servers?

In some web applications, when a user make a GET request for url1, there is a high probability of him requesting a url2 within the next few seconds/minutes. Especially For some applications with fixed operation path. In this case, it makes sense to preload the result of url2 for the user and cache it, to reduce the load time for url2.

I am wondering if there is any open-source solution that solve this problem?

1. Record a user's requests history, and predict his next request(s) based on current request. It's better if it can accept hard configuration as well as learn from history.

2. Based on the probability of prediction, and current system load, to decide whether to preload the next url.