new icn messageflickr-free-ic3d pan white
Scaling Quantum Computing: 17 Years of Rose's Law | by jurvetson
Back to photostream

Scaling Quantum Computing: 17 Years of Rose's Law

Today D-Wave announced the world's first quantum computer with over 5,000 qubits (quantum bits). Qubit count is one of the metrics of a quantum computer's computational power, which grows dramatically with the number of qubits. I have been tracking the frontier of quantum computing from the beginning, and this is the updated semi-log graph (where a straight line is an exponential curve).


When I first invested in quantum computing in 2003, D-Wave founder Geordie Rose had one working qubit, and he predicted that he would be able to demonstrate a two-bit quantum computer within 6 months. There was a certain precision to his predictions. He went on to suggest that the number of qubits in a scalable quantum computing architecture should double every year. It sounded a lot like Gordon Moore’s prediction back in 1965, when he extrapolated from just five data points on a log-scale. So, I called it “Rose’s Law” and that seemed to amuse him.


In 1965, the original Moore's Law paper predicted an annual doubling of transistor counts, but ten years later, Moore revised his “law” to every 24 months. With a little hand waving, most reports today attribute 18 months to Moore’s Law, but there is quite a bit of variability.


Rose’s Law bears an uncanny resemblance. It followed an annual doubling for the first decade, and now seems to be following a biennial doubling over the past seven years to the present day. Now, let me caution that adding a segmented overlay analysis to a scatter plot is ripe for interpretive bias. There is no specific reason for 2013 to be a breakpoint in the slope. We could alternatively plot a best-fit line through all 17 years, and it would have roughly an 18-month doubling time. So, time will tell. With a few more product releases from D-Wave, we’ll see how closely Rose’s Law recapitulates Moore’s Law in the quantum computing domain. D-Wave has been the dramatic leader in qubit count throughout the 17 years, much like Intel was in the early days of Moore’s Law (before handing the baton to NVIDIA).


With today’s 5K qubit announcement (the quantum Advantage computer), D-Wave’s CEO Alan Baratz emphasized their leadership position and scalability:


"There is no other quantum computer anywhere in the world that can solve problems at the scale and complexity that this quantum computer can solve problems. It really is the only one that you can run real business applications on." — VentureBeat


“With over twice the number of qubits, with over twice the connectivity, with over five times the number of devices on the superconducting chip, we’re still able to program it in the same amount of time, read out in the same amount of time, run it at the same temperature — which means we’re able to continue scaling the technology. This is a really important point because, over the years, there have been various so-called experts who have said that the D-Wave technology just wouldn’t scale. And yet, we’re the only quantum computing technology that has scaled.”



And a more technical dive in Ars Technica: "What's it take to make a chip with over a million Josephson junctions?" and details on the D-Wave quantum annealer and programing model.


Full disclosure: I served on the board for the first 17 years, and shared a lot of photos here on flickr

5 faves
Uploaded on September 30, 2020