Exactly what "core" are you using?
What chip is it in?
What speed are you running it at?
What work have you done to identify bottlenecks, inefficiencies, etc, in your existing system?
etc, etc.

Many modern derivatives use less clock cycles per instruction and can be clocked at higher frequencies - so can be much faster than old variants.

However, it does sound like your application is excessive for an 8051...