PORTLAND, Ore.—Micro-electro-mechanical systems (MEMS) moved closer to enabling quantum computing with university researchers demonstrating that micro-mirrors can read and write qubits encoded on clouds of ultra-cold atoms suspended in a transparent media.
Semiconductor memories today need bit-lines to address them before reading or writing, but according to Duke University and the University of Wisconsin-Madison, qubits can be likewise addressed with two lasers focused on them by MEMS micro-mirrors.
In the demonstration setup, clouds of five rubidium-87 atoms were spaced at 8.7 micron intervals and addressed by two lasers which independently targeted their location, potentially allowing qubits to be written and read out by lasers. The current experimental setup merely proved the concept, but future quantum computers could use a transparent media to store qubits close enough to each other that their interactions could perform ultra-complex calculations that are intractable today—such as cracking long encryption codes.
Ultracold cloud of atoms used to store quantum qubits in the laboratory of Mark Saffman's group at University of Wisconsin-Madison.
An access time of about five microseconds, to switch between qubits, was reported by the researchers to be about 1,000-times faster than today's micro-mirrors used in optical switches. Next the group plans to construct what they believe will become the basic building block for future quantum computers--two-qubit gates confined in planar two-dimensional arrays.
I agree--that picture does look a lot like HAL. The line between preposterous science fiction and realizable technologies is so blurred that its increasingly hard to tell the nuts from the geniuses anymore!
I keep remembering the transparent storage media in the movie 2001--remember when Dave was unscrewing them and they popped out looking like bars of glass. How else would the memory in those bars be accessed if not by high-precision laser targeting? I think many of us guessed that MEMS micro-mirrors would be the enabler for storing qubits on atoms in a transparent medium, but Duke and Univ. of Wisconsin were the first to prove the concept.
MEMS micro-mirrors are already used for optical switching by companies like Glimmerglass Networks and Micralyne, but those devices do not have the speed or accuracy necessary to create a quantum computer. The Duke and the University of Wisconsin demonstration, however, shows that MEMS is up to the task of targeting individual atoms, albeit it will be a long time--at least a decade--before we see this dream fully realized.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.