MOORE’S LAW AND MICROPROCESSING POWE

: Introductionut

A microprocessor incorporates the functions of a computer's central processing unit (CPU) on a single integrated circuit

A microprocessor is a multipurpose

 device that accepts digital data as input, processes it according to instructions stored in its memory, and provides results as output.

The microprocessor is the heart of any normal computer, whether it is a desktop machine, a server or a laptop.

- The microprocessor contains all, or most of, the central processing unit (CPU) functions and is the "engine" that goes into motion when you turn your computer on. A microprocessor is designed to perform arithmetic and logic operations by using registers.

Microprocessors are used in electronic devices to compute data. They can be found in just about any modern electronic device, from televisions to personal computers.

MOORE’S LAW AND MICROPROCESSING POWE

In 1965, Gordon Moore, the director of Fairchild Semiconductor’s Research and

Development Laboratories, an early manufacturer of integrated circuits, wrote in Electronics magazine that since the first microprocessor chip was introduced in 1959, the

number of components on a chip with the smallest manufacturer costs per component

(generally transistors) had doubled each year. This assertion became the foundation of Moore’s Law.

Moore later reduced the rate of growth to a doubling every two years

(Tuomi, 2002).

This law would later be interpreted in multiple ways. There are at least three variations of Moore’s Law, none of which Moore ever stated: (1) the power of microprocessors doubles every 18 months (Tuomi, 2002); (2) computing power doubles every 18

months; and (3) the price of computing falls by half every 18 months.

Figure 4-3 illustrates the relationship between number of transistors on a microprocessor and millions of instructions per second (MIPS), a common measure of processor power. Figure 4-4 shows the exponential decline in the cost of transistors and rise in

computing power.

There is reason to believe the exponential growth in the number of transistors and the

power of processors coupled with an exponential decline in computing costs will con-

                                tinue into the future.

 Chip manufacturers continue to miniaturize components. Intel has

recently changed its manufacturing process from 0.13-micron component size (a micron

is a millionth of a meter), introduced in 2002, to a newer 90-nanometer process in 2004

(a nanometer is a billionth of a meter). With a size of about 50 nanometers, today’s transistors should no longer be compared to the size of a human hair but rather to the size of

a virus, the smallest form of organic life.

By using nanotechnology, Intel believes it can shrink the size of transistors down to

the width of several atoms. Nanotechnology uses individual atoms and molecules to create computer chips and other devices that are thousands of times smaller than current

technologies permit. IBM and other research labs have created transistors from nanotubes and other electrical devices (IBM, 2004) and have developed a manufacturing

process that could produce nanotube processors economically (Figure 4-5). Other new

technologies include strained silicon, 300-millimeter production wafers (which decrease

the costs of production), and denser interconnections among components.

Whereas the first Pentium microprocessors operated at 75 megahertz, today’s

Pentiums are available with 3-gigahertz speeds. However, increasing processor speeds at

the same exponential rate as in the past may no longer be possible. As processor speeds

increase, heat is generated that cannot be dissipated with air fans.

Another brake on future increases in microprocessor speed is more market-oriented:

Most consumers may not need vast increases in microprocessor speed but instead are

more interested in low power consumption for longer battery life and low weight to

increase laptop and handheld computer portability. For this reason, Intel and other firms

are designing the next generation of chips to be less power hungry and lower in weight

even if they are the same or even slower speeds. Other options include putting multiple

processors on a single chip

Moore’s Law and the Future of Microprocessor

Technology 

 "Moore's Law and the Future of Microprocessor Technology" looks at IBM's Blue Water Supercomputer that is slated to be installed in 2011 as breaking ground for new microprocessor technology. While current microprocessor manufacturing methodology is about to reach its limits, nanophotonics looks like a viable breakthrough technology that in the near future could show up in consumer PCs.

ZDNet’s Jason Perlow wrote an interesting article concerning recent breakthroughs in microprocessor technology. Moore’s Law has been a central principle in the evolution of current microprocessor technology, simply stating that the number of transistors that can be placed inexpensively on an integrated circuit has doubled approximately every two years, and will continue to do so in the near future. Moore’s Law has been accurate for the last half century. However, Jason Perlow has described a recent development at IBM that will likely put Moore’s Law on steroids.

For those of you interested in reading Jason’s blog, here is the link to his article on ZDNet. Jason writes, “instead of just proving itself consistently correct, Moore’s Law is going to have to be completely re-written — instead of microprocessor technology doubling its performance every two years, we’ll be looking forward to ten to twenty fold increases in computational power, at a bare minimum, every five years.” 

The reason for Jason’s assertion is that IBM unveiled on December 1st of this year a technology it calls CMOS Integrated Silicon Nanophotonics. Rather than utilizing semiconductor pathways, as is currently the case, data is routed for processor interconnects via light pulses instead. 

Currently, advances in microprocessors are realized reducing the size of transistors to cram more of them onto a piece of silicon. Today, the reductions in transistor size has reached 32 nanometers (We are talking very small here), and it is generally expected that the processes in use to accomplish these reductions will bottom out at 11 nanometers, which can reasonably be expected to happen in just a few short years. 

IBM’s CMOS Integrated Silicon Nanophotonics technology will initially be used in the Blue Waters Supercomputer that is scheduled to be installed in the National Center for Supercomputing Applications in Urbana, Illinois in the summer of 2011. Currently, the most powerful supercomputer in the world is the Chinese Tianhe-1A, which has achieved 2.67 Petaflops or 2.67 quadrillion floating-point operations per second. I cannot even begin to relate to that kind of computing power. The Tianhe-1A was constructed on the current microprocessor technology. To put things in perspective, the IBM Blue Waters Supercomputer is expected to be four times faster than the Tianhe-1A, and this is just in the infancy of this technology! The maturing of this technology will be mind-boggling. The impact on the fields of science and medicine alone are difficult to imagine.

What does this mean for consumers? Eventually this new technology will filter its way down to Enterprise and then to consumer computing. In the near future, we can expect to see personal computers that make our current top of the line units seem like clay tablets in comparison. 

This is something to continue watching, and I am certain we will be hearing 

more about it in the months to come. 

What are your thoughts? Any comments are encouraged and welcome