Pollack's rule

from Wikipedia, the free encyclopedia

The Pollack's Rule , named after Fred Pollack a developer of the company Intel , is a rule of thumb from the processor architecture . The rule states that an increase in the complexity of a chip by the factor only leads to an increase in the computing power by the factor . A doubling of the chip complexity ( ), for example in the form of a multi-core processor in which two processor cores of the same type are used instead of just one processor core, increases the computing power of the overall system by an average of only 41%.

Pollack's rule is not a scientific law of nature , but a rule of thumb based on empirical observation. Moore's law represents a similar empirical relationship , but in contrast to Pollack's rule, it relates development to the temporal course of technological progress. The technological progress or complexity of a chip is expressed in both cases in different key figures. Typical and common for this are the reduction of the structural sizes of microchips or the number of transistors on a microchip.

Individual evidence

  1. Fred J. Pollack: New microarchitecture challenges in the coming generations of CMOS process technologies (keynote address) . In: Proceedings of the 32nd annual ACM / IEEE international symposium on Microarchitecture . IEEE Computer Society, Washington, DC, USA 1999, ISBN 0-7695-0437-X ( presentation slides [PDF; 281 kB ]).
  2. Shekhar Borkar: Thousand core chips-A Technology Perspective . Ed .: Intel Corp, Microprocessor Technology Lab. ACM / EDAC / IEEE Design Automation Conference, 2007 ( Online [PDF; 224 kB ]).
  3. Shekhar Borkar, Andrew A. Chien: The future of microprocessors . In: Communications of the ACM . tape 54 , no. 5 , 2011, p. 67-77 , doi : 10.1145 / 1941487.1941507 .