Scientists at the University of California at Berkley, believe that in the near future, computers will compute with nanometer-sized bar-magnets (nano-magnets) resulting in one million times less energy being required than current silicon-chip-based computers use.
This is possible because no electrons will be used for the computers logic operations. No electrons means…no electrical resistance and no waste heat. We’ve all had hot laptops on our laps. That and much more could become a thing of the past.
You may think that power requirements that are one millionth what they are now is impossible and you’d be almost correct.
This is because in actuality this is right at the limit of what physics allows. There’s even a term for this exact situation; it’s called the Landauer limit. That’s the the absolute minimum possible energy required to change one bit of information.
50 years ago IBMer Rolf Landauer had his hands on a new tool called information theory. He used this theory to calculate the minimum energy required for a logic operation like an AND or an OR operation.
The 2nd law of thermodynamics says that entropy or disorder always increases in a closed system right? That’s why perpetual motion machines are bunk.
In this context the law says that a process that’s irreversable like a logical operation or the erasure of a bit of information — dissipates energy that cannot ever be recovered. The Landauer Limit is that minimum unrecoverable amount of energy that physics absolutely requires for these operations.
The magnets themselves are 100 nanometers wide and 200 nm long. They’re like bar magnets with a north-sout polarity. The ones and zeroes of binary code are then represented by the way the pole is oriented. The interactions of these different nanomagnets then permit logic operations to be carried out.
If practical, this technology would be revolutionary for many reasons. Imagine battery life being extended to an unheard of degree like running a laptop for 10 years on a single charge…I’m just throwing that number out there.
Power consumption and heat generation are another big concern and this is especially true the faster and more densely packed computers become. As we transition from petascale supercomputers to exascale (one quintillion flops and 100s of millions of cores), the power required would actually untenable with current designs.
For example: For the fastest supercomputer on the planet, China’s Tianhe-1A, to achieve exascale performance, it would require more than 1.6 GW of power – an amount large enough to supply electricity to 2 million homes.
Jeffrey Bokor, UC Berkeley professor of electrical engineering and computer sciences said:
“In principle, one could, I think, build real circuits that would operate right at the Landauer limit, Even if we could get within one order of magnitude, a factor of 10, of the Landauer limit, it would represent a huge reduction in energy consumption for electronics. It would be absolutely revolutionary.”