Beyond Moore’s Law

I remember many articles in 2016 that lamented that Moore’s Law was dead, which spelled the end of the era for U.S. technology constantly improving due to ever-faster generations of computers. Moore’s Law is named after Gordon Moore, an engineer who later became one of the founders of Intel. In 1965, Moore observed that the number of transistors that could be squeezed into a given area of a circuit board was doubling every two years. He predicted this trend would last for perhaps another decade, but the microchip industry fulfilled his prediction for over 50 years.

In 1965 a single transistor cost about $9 in today’s dollars and we can now put billions of transistors onto a chip, at a tiny fraction of a cent each. The belief that chips could always be improved helped to launch Silicon Valley and enabled the huge array of technological changes that have been brought about by cheap computer chips. The companies that make chips thrived by creating a new generation of chips every few years, which represented a significant leap forward in computing power.

It’s now clear that Moore’s law is dead, or nearly so, and computing is not going to improve much from denser transistors. But that doesn’t mean that computing can’t get faster, and there are new strategies for developing better computers and chips.

One path is to use new materials and devices that can improve the computing process.

  • Researchers have been exploring new materials for chips such as graphene and carbon nanotubes.
  • Scientists are pursuing optical chips that use light for processing inside the chip instead of electricity. This speeds up the bottleneck of getting data into and out of a chip.
  • Spintronics research is looking at using the characteristics of the spin direction of electrons as a way to create much higher density data storage.
  • Tunnel field-effect transistors switch between 1s and 0s using quantum tunneling instead of today’s modulation with thermionic emissions. This has the potential to create chips that use far less power.

Another path for future improvements is to develop new models of computing.

  • Quantum computing is exploring the ability to process multiple calculations simultaneously.
  • Neuromorphic computing models the computing process after systems in the human brain and nervous system.
  • Adiabatic computing uses reversible circuits that have as many outputs as inputs. Since each input can be reconstructed from an output, no bits are lost, and reversible circuits give off no heat.

Another approach is to develop a more efficient architecture and packaging of chips.

  • There have been some significant improvements in building three-dimensional chips that consist of stacked layers of chips.
  • Reconfigurable computing is an architecture that can speed up complex processing by using components that can change function or spatial configuration during the computing process.
  • Dark silicon computing powers down any portion of a chip that is not being used to conserve power.
  • Superconducting computers operate at cold temperatures that dramatically cut power usage and speed up calculating.
  • Near-threshold voltage computing saves power by using chips that operate only at the peak energy-efficient level.

The biggest bottleneck today for creating the next generations of better computing is the typical 10-year window required to go from an idea created in a lab to producing chips. If we want to continue on the path of predictably better computers, we’ll have to find a way to speed up this process.

New Technology – February 2017

grapheneThere has been so much going on in the telecom industry lately that I haven’t published a blog examining promising new technologies for a while. Here are a few new breakthroughs that ought to eventually affect our industry:

Metal that Conducts Electricity but not Heat. Physicists at the Lawrence Berkeley National Lab and UC Berkeley have found a metal that contradicts the Wiedermann-Franz Law.  This Law states that good conductors of electricity will also be proportionately good conductors of heat. The physicists were working with vanadium dioxide and unexpectedly discovered this property. There are a few other materials that are much better at conducting electricity than heat, but they only do so at temperatures a few hundred degrees below zero. It appears vanadium dioxide can do this at room temperatures. This property is derived from the fact that electrons move through the metal in a synchronized manner which is normally observed only in fluids, instead of individually which is normally observed in metals.

There is great potential for a material with this property – it could be used as an insulator in computers to keep components cool and to drastically lower the cooling costs experienced in data centers. On a more macro level this could lead to better insulation in homes and appliances and could drastically improve energy efficiency in a wide range of applications.

Superconductor Graphene. Researchers at the University of Cambridge in the UK have found a way to induce superconductivity in graphene. Today all superconducting materials only function at temperatures below -454 degrees Fahrenheit. But their research indicates superconducting graphene will work at much higher temperatures. The researchers created superconducting properties by layering graphene only on an underlying sheet of metal.

Superconduction is a big deal, because in the ultimate state a superconductor passes electrons with zero resistance. Compare that to normal materials, such as our electric grid that loses 7% of generated power getting to homes, and the difference is remarkable.  Finding a room-temperature superconductor would be a huge breakthrough because it could mean electric transmissions with no power losses and an end to the heat generated in electronics and appliances that comes from resistance.

Mass Producing Graphene. Scientists at Kansas State have found a cheap way to mass produce graphene. They discovered the process when working with carbon soot aerosol gels. The process is simple and only requires hydrocarbon gas, oxygen and a spark plug. The gases are forced into a chamber and graphene is formed with a spark. This is a low-power way to make graphene since it only needs a spark rather than continuous power.

Until now graphene has been expensive to make in quantities greater than milligrams and the process required caustic chemicals. With this method it’s easy to make graphene in gram quantities and the process ought to be scalable to much larger quantities.

Better Use of Wireless Spectrum. Engineers at UCLA have found a technique that might allow better use of wireless spectrum. They have found a way to use a tiny device called a circulator that allows a chip to use both incoming and outgoing signals of a given spectrum at the same time. Today’s technology only uses spectrum in one direction since dual use of spectrum has caused interference.

Circulators have been tried before, but earlier devices used magnetic materials which can’t be incorporated into chips. The prototype they built uses coaxial cables to route the signals through non-magnetic materials and they believe the design can be built directly into silicon.

The circulator works by sequentially switching signals using different paths in a similar manner that a busy train station can have trains coming in going in both directions. The design uses six transmission lines and five switches which are turned off and on sequentially to allow incoming and outgoing signals to pass each other without interference.

This would be a big breakthrough for cellphones since it would allow for better use of the spectrum. This wouldn’t increase data speeds, but would allow a cell site to handle more phones at the same time.

New Technology – June 2016

The InternetThere is a lot of recent news of technological breakthroughs that ought to have some an on telecom and broadband.

Faster Microwave Radios. A collaboration of researchers working for ACCESS (Advanced E Band Satellite Link Studies) in Germany has created a long-range microwave link at 6 Gbps speeds. The technology uses the very high E band frequencies at 71 – 76 GHz and in testing were able to create a data path between radios that were 23 miles apart. It’s the very short length of the radio waves at this frequency that allow for the very fast data rates.

The radios rely on transistor technology from Fraunhofer IAF, a firm that has been involved in several recent high-bandwidth radio technologies. The transmitting radio broadcasts at a high-power of 1 watt while the receivers are designed to detect and reconstruct very weak signals.

When perfected this could provide a lower cost way to provide bandwidth links to remote locations like towns situated in rough terrain or cellular and other radio towers located on mountaintops. This is a significant speed breakthrough for point-to-point microwaves at almost six times the speed of other existing microwave technologies.

Smarter Chip Processing. A team at MIT’s Computer Science and Artificial Intelligence Laboratory have developed a programming technique to make much better use of denser computer chips. In theory, a 64-core chip ought to be nearly 64 times faster than one with a single core, but in practice that has not been the case. Since most computer programs run sequentially (instructions and decision trees are examined one at a time, in order) most programs do not run much faster on denser chips.

The team created a new chip design they call Swarm that will speed up parallel processing and that will also make it easier to write the code for denser chips. In early tests, programs run on the swarm chip have been 3 to 18 times faster while also requiring as little as 10% of the code needed for normal processing.

1,000 Processor Chip. And speaking of denser chips, scientists at the University of California at Davis Department of Electrical and Computer Engineering have developed the first chip that contains over 1,000 separate processors. The chip has a maximum computation rate of 1.78 trillion instructions per second.

They are calling it the KiloCore chip and it’s both energy efficient and the highest clock rate chip ever developed. The chip uses IBM’s 32 nanometer technology. Each chip can run separate programming or they can be used in parallel. The scientists envision using a programming technique called the single-instruction-multiple-data approach that can break applications down into small discrete steps so that they can be processed simultaneously.

Faster Graphene Chips. Finally, US-Army funded researchers at MIT’s Institute for Soldier Nanotechnologies have developed a technology that could theoretically make chips as much as 1 million times faster than today. The new technology uses graphene and relies on the phenomenon that graphene can be used to slow light below the speed of electrons. This slowed-down light emits ‘plasmons’ of intense light that create what the scientists called an optic boom (similar to a sonic boom in air).

These plasmons could be used to greatly speed up the transmission speeds within computer chips. They have found that the optic booms can push data through graphene at about 1/300th the speed of light, a big improvement over photons through silicon.

The researchers have been able to create and control the plasmon bursts and are hoping to have a working graphene chip using the technology within two or three years.