. . "The term applies especially to systems that function above a teraflop or 10^12 floating-point operations per second, and is also often used as a synonym for supercomputing." . .