There are still many places in the code where we use System.currentTimeMillis() to measure time intervals. This method is known for its lack of acucracy and should be replaced with a call to nanoTime() wherever possible, as it provides a sub-microsecond accurracy on most, if not all, modern hardware.

This should also be the opportunity to review the parts of the code that rely on waiting for a specified time, i.e. wait(...), sleep(...) and wrapping methods in class ThreadSynchronizaion and others, where there is a potential for blocking threads for a large amount of time and generate deadlock-like situations.