One of the key limitations to a single-core application running on a machine is the clock speed of the CPU. When we talk about clock rate, we are essentially talking about how many clock cycles a CPU can execute every second.
For the past 10 years, we have watched as manufacturers managed to surpass Moore's law, which was essentially an observation that the number of transistors one was able to place on a piece of silicon doubled roughly every two years.
This doubling of transistors every two years paved the way for exponential gains in single-CPU clock rates, and CPUs went from the low MHz to the 4-5 GHz clock speeds that we now see on Intel's i7 6700k processor.
But with transistors getting as small as a few nanometers across, this is inevitably coming to an end. We've started to hit the boundaries of physics, and, unfortunately, if we go any smaller, we'll start being hit by the effects of quantum tunneling. Due to these physical limitations, we need to start looking at other methods in order to improve the speeds at which we are able to compute things.
This is where Materlli's Model of Scalability comes into play.
