A method was developed to make queries faster in Internet databases by adapting to real-time conditions. The challenge lies in optimizing queries over faraway servers via the Internet, as server speeds and network congestion can vary. Traditional methods sometimes choose expensive query plans by mistake due to unreliable estimates and not considering certain costs. By using a probing technique, they detected and switched to cheaper plans during query execution, reducing unnecessary work. The study showed that this adaptive approach outperformed static optimization, especially for join queries. This method could also help choose the best locations for executing specific functions in a client-server setup. The implementation in Java showcased the effectiveness of this real-time optimization approach for improving query performance in distributed databases.