User:Vipul/Dennard scaling

Dennard scaling is a loose term based on a 1974 paper co-authored by Robert H. Dennard, after whom it is named. Originally formulated for MOSFETs, it states, roughly, that as transistors get smaller their power density stays constant, so that the power use stays in proportion with area: both voltage and current scale (downward) with length.

Relation with Moore's law and computing performance
Dennard scaling relates Moore's law (which postulates a reduction in the size of transistors leading to more and more transistors per chip at the cost-effective optimum) and the claims that the performance per watt of computing is growing exponentially at roughly the same rate. This is closely related to Koomey's law, which says that performance per watt in computing has been doubling every 1.57 years (somewhat faster than the doubling period of Moore's law, which is about 1.8 years).

Recent breakdown of Dennard scaling
Since around 2005-2007, Dennard scaling appears to have broken down, so even though Moore's law continued for several years after that, it has not yielded dividends in improved performance. The primary reason cited for the breakdown is that at small sizes, current leakage poses greater challenges, and also causes the chip to heat up, which creates a threat of thermal runaway and therefore further increases energy costs.

The breakdown to Dennard scaling prompted a switch among some chip manufacturers to a greater focus on multicore processors, but the gains offered by switching to more cores are lower than the gains that would be achieved had Dennard scaling continued.