COMPUTING BEYOND MOORE’S LAW

Article Review Example: COMPUTING BEYOND MOORE’S LAW

Bibliographic Reference

Shalf, J.M., & Leland, R., (2015). Computing beyond Moore’s Law. Computer, 48(12), 4-23. doi:10.1109/MC.2015.374

Objectives

• The author explains Moore’s theory, an economic theory rather than a physical law, while exploring the correlation between the power used by circuits and their size.
• The purpose of this article is to shed light on the future of circuits as computing technology advances.
• The author communicates to his audience that Moore’s law is coming to an end.
• This article gives input on what must be done in the future to ensure growth within the software industry.
• The author wanted the reader to understand that there is a huge potential for innovation within the IT market as it pertains to this field.
• This article explores new computing models that could prove useful to devices while circuits are scaled onto the atomic level.

 

Summary

In the 1960s, Moore and Dennard had theories about computing technology. Moore’s theory was that every year, the components on an integrated circuit would double. His theory was correct, and it lasted 40 years longer than he assumed it would. Dennard’s theory was that if one shrinks the components of a circuit, and it is processed with the same amount of energy as before, it will be processed more quickly because it is smaller. His theory was also known as Dennard scaling. These computing technologies were applicable when circuits were measurable with standard equipment. However, now circuits are being created on the atomic and nanoscale level. Because of this, scaling will no longer be conventional, and Dennard’s and Moore’s theory will soon come to an end.

As the scale of circuits is reduced to nanometers, efficiency in energy, power, and storage will all drastically change as well. Unless new technology is created, advancements in the correlation between the energy and capacity of a circuit will be at a standstill; this will be because of the lack of mutual reinforcement of hardware and software within technology. The next components of semiconductors and microelectronics have not yet been created, but this article suggests three ways in which this technology can be discovered: through the production of new equipment, the construction of new architecture regardless of the availability of new computational devices, or the augmentation of computing models. Ground-breaking research and innovation in this area shows high promise because of the annual $4 trillion market prodding movement into unexplored ideas in the field of information technology.

New practical devices created to process atomic-level technology must be based on advanced computational approaches. These devices, or at least their components, should be scalable. This will allow them to be used in a variety of different industries because progress is driven by the scalability of technology. Atomic-particle-technology must also show a promising future in computational abilities for it to be seen as valuable for the future.

New computational paradigms can be created by looking back and expanding upon technology that is not fitting for current computational devices using CMOS logic. A strategic foundation for usage must be laid now so that the 10-20 year implementation process can begin as soon as possible. Even with the development of new technology, it must be able to be implemented in an affordable manner on a small scale to make it economically practical. All this advancement is simply about pushing limits as an intellectual society. Moore’s theory was based on 2D lithography. The race for nanotechnology is on because photolithography is largely predicted to reach its limit in 2020.

Results

• The end of Moore’s law has huge implications for the future of consumer electronic devices
  • Companies depend on a relatively constant and steady increase in the functionality of technology for their products to increase storage or battery life.
  • This threatens the growth of consumer-focused industries while waiting for the creation of new atomic and nanotechnology.
• Creating new architectural patterns can broaden computing technology.
  • This new technology can be created by rearranging the basis of current computing technology without actually creating new technology.
  • Architectural schemes can help manage energy, battery usage, hardware cost, and improve bug detection.
• In an effort to lower voltage for the creation of new devices, one can rewire the circuitry.
  • There is potential for the voltage to be lowered 2 to 3 times simply based on design.
• Heterogeneous semiconductors may be competition for homogenous silicone semiconductors within devices that will be able to compute atomic particles.
  • Heterogeneous semiconductors have stayed out of the mainstream because they are hard to manufacture although they show great promise in their performance.
  • However, integration of column III-V materials is now possible because of dramatic improvements in research.
• There are computing devices based on the brains of animals.
  • These are highly complex machines. Moore’s law ensured that those using biological machines could acquire materials with high processing speeds at a low cost, but now the materials needed for this type of machine is at risk.

 

Worldview Consideration–Ethical or Legal Considerations

Comment on any ethical or legal considerations this lab may have enlightened you about.  If none were obvious during the lab, then review the lab and identify any ethical or legal concern that comes to mind. This must be in paragraph form.  Bullets are not acceptable.

Worldview Consideration–Christian Worldview

Comment about how the technology revealed in this lab intersects with theology.  This must be in paragraph form.  Bullets are not acceptable.

Questions

1. What is the difference between 2D and 3D Lithography?

The energy needed to move information is proportional to how far it must go. Data movement will be in control of energy losses and it will be virtually impossible to continue creating circuits vertically which will lead to the creation of 3D circuits.

2. How does one go about creating heterogeneous silicon semiconductors?

Silicon straining is a process that “alters the silicon substrate so that its atomic spacing aligns with that of the III-V material and dopes the III-V materials with additional impurities so that the atomic spacing aligns with that of silicon when it is vapor-deposited onto the silicon substrate”(Shalf 2015). This allows heterogeneity to be possible.

3. What does successful implementation of this advanced computing technology look like in five years?

I have concluded that there will be a lot of research and development in the search for new computing technologies. The deadline is about 2020, before we run out of innovation possible with current technology. Many people must be currently researching new methods, so ideally in five years, a new architecture and framework for a technology that can process atomic-level circuits will be built and begin to be implemented.

 

 

References

Only needed if references are used.

Last Name, F. M. (Year). Article Title. Journal Title, Pages From – To.

Last Name, F. M. (Year). Book Title. City Name: Publisher Name