In our pocket we carry a machine unimaginable a few years ago, capable of finding the information we want in a few seconds.
The next challenge is to minimize your energy expenditure. The work of physicists like Albert Fert, discoverer of the giant magnetoresistance, lays the foundations to face the current energetic hemorrhage caused by our searches on Google.
If we had the ability to design the ideal computer or mobile, what would it be like? Many users would focus on aesthetics, others on functionality or originality. But in reality, physicists like Albert Fert (Carcassone, France, 1938), are concerned about storage capacity, power consumption, and battery life.
Several companies are already in the process of manufacturing these mobiles that can last from seven to fifteen days without being recharged.
“Afterwards I don't have a particular idea of what that mobile of the future will be like”Confesses the discoverer of the giant magnetoresistance (1988), an effect of quantum mechanics that has made it possible to miniaturize the readers of the hard drives of computers and increase their storage capacity.
This gave rise to spintronics, a technology that manipulates the quantum state of particles - the spin of the electron - for applications in the computer industry. Thanks to its properties, the result will be mobiles, computers and hard drives, among others, more efficient, faster and with greater storage capacity.
“20% of the world's electricity consumption in 2030 will come from digital data transmission"Says the physicist, recalling a study published in the journalChallenges.
What does searching on Google mean?
Thedata centers or data processing centers, spaces where all information is managed by large companies and organizations, are undoubtedly the largest consumers of energy, even more than personal computers themselves.
“As more and more things pass through the internet, the energy consumption of data centers will increase in the next ten years by a factor of 10”Says the French physicist, who emphasizes that a Google data center in the US uses as much energy as the city of San Francisco.
But what does the energy we spend browsing the net equal to?
“One piece of information that I always give is that thirty searches on Google correspond to the energy needed to boil a liter of water”, Fert points out. With this data it is not difficult to extrapolate the huge amount of electricity that is consumed, and that directly affects the environment, just to keep us informed.
Fert has it clear: "We spend uselessly when we search the internet. With a computer we could find it directly but, out of laziness, we go back to Google and create links to get to a website”. Not even renewable energies, such as solar and wind, are currently capable of offsetting the energy consumption of computers.
Less energy and more space
The French Nobel Prize raises a question regarding this problem: “How to continue improving information technologies while slowing down this hemorrhage in energy consumption?”. For this, a new generation of computers will soon arrive that will consume much less and will be based on the effects of quantum mechanics.
“Currently, in computers, what consumes energy is RAM, the data storage that to fulfill its work has to be continuously providing energy”Declares the physicist.
In recent years the development of spintronics has accelerated in multiple directions, not only culminating in hard drives and STT-RAM products.
A computer that mimics the brain
One of the applications is computers and neuromorphic systems, that is, computers inspired by the brain. "This is starting”, Emphasizes the physicist. Today computersbioinspired they reproduce synapses and neurons with microcircuits.
“But it takes hundreds of transistors to make a neuron and dozens to make a synapse, and the result is huge computers that also consume a lot of energy like those in the Alphago program, which use 10,000 times more energy than the Go player in front of you.”, Specifies French.
These machines use a different type of calculation, rather neural, to that used by usual computers, mainly digital. In this case, they have specific compounds based on other nanophysical materials.
But in the brain, the process differs: "Memory and signal transfer to memory occur at the same time. As soon as a signal passes from neuron to neuron, each time a synapse is crossed, it changes and is memorized”Says the physicist. Analog signals from the brain allow more nuanced vision that is transmitted by millions of synapses and neurons.
Will it ever be possible to use computers that mimic the brain? For this, a synapse would have to be made with 50 transistors and a neuron with 500. "This is huge. If we wanted to do something with brain power we would need a computer the size of Madrid”, He confesses.
This technology, specially developed by French researcher Julie Grollier, with whom Fert has worked, is still limited. "The components and effects that mimic neurons and synapses do not yet exist", He says. "We are only at the beginning”Concludes the Nobel Prize.