The company based in North Carolina reported Thursday that it had created the fastest Quantum Computing system globally, at least twice as fast and powerful as the current Google and IBM computers.
Honeywell claimed that it improved the efficiency of its Quantum Computers enough to beat that of IBM’s by one closely monitored measure. The achievement gives Honeywell’s computer the title of the world’s fastest quantum computer, at least for now.
The achievement meets a promise made by Honeywell in March. It is a breakthrough in the company’s ambitious attempt to increase its quantum computing.
This consists of an inox steel chamber of the size of a basketball, which is refracted by liquid Helium at a temperature just above absolute zero, the level at which the atoms stop vibrating. The system is in a 1500 m2, high-security storage area in Boulder, Colorado. Within that space, lasers are used to measure individual atoms floating over a computer chip.
The quantum computer of Honeywell, nicknamed H0, scored 64 on a metric known as quantity. This test tests both the total number of qubits of the machine – the fundamental components processing data in a quantum computer – and the machine’s use of these qubits. In January, the result was higher than the 32 IBM system Raleigh.
While experts have been studying the capability of quantum computing for decades, that is, designing machines with the ability to perform calculations beyond the limits of supercomputers and classic computers, the industry has been restricted to the intrigue of research teams at computing companies such as IBM and Google, until recently.
But the race between those companies over the past year to claim supremacy and provide a commercial use — has heated up in the quantum computer race. Honeywell’s machine has achieved a Quantum Volume of 64, a metric designed by IBM that measures machine capability and error rates. Still, it is also difficult to decipher (and potentially game possible, as the quantum computing expert Scott Aaronson wrote in March). By way of comparison, IBM announced in January that its newest machine, Raleigh, had achieved a Quantum Volume of 32.
Google also spend massive resources trying to develop its quantum capacities, and, in October, it had created a machine that would have required a supercomputer to process in only 200 seconds for a period of 10,000 years. (IBM refuted Google’s claim, saying it only took 2,5 days to finish the calculation.)
To find solutions that are practically impossible to standard “classical” computers, Quantum Computing focuses on the weird rules of atomic physics. This is the guarantee, at least. In the past, a quantum computer is mostly finished research ventures, with classical computers only outperforming in one tiny Google’ quantum dominance’ study.
But as researchers add additional qubits and study how to defend them from distractions that disrupt calculations, quantum computers are expected to be more durable. Tasks that are expected to be successful include improving financial portfolios, learning machinery, and developing new materials such as solar panels or batteries.
In the last decade, Honeywell developed technologies to produce cryogenics and laser techniques to achieve this objective. The company has built up a team of more than 100 experts in the past five years that has entirely built up the machine, and Honeywell confirmed it would take three months to complete in March. It could achieve a goal even if the Covid-19 reversed its workers and forced some employees to work remotely.
“We had to redesign how we work in the facilities completely, had to limit who was coming on the site, and put in place physical barriers,” says Tony Uttley, president of Honeywell Quantum Solutions. “All of that happened at the same time we were planning on being on this race.”
Quantum computing is taking a fresh start.
Honeywell was a leading manufacturer of massive mainframes decades ago, but he sold the business and left the computer industry to other vendors. It has quietly developed and used its experience in materials science and industrial activities to deal with quantum technology in recent times. It is a group of 120 experts in Colorado and Minnesota.
The current design of Honeywell, H0, reaches 64,00 to 640,000 from its new quantum number, adding more qubits, said Honeywell’s president Tony Uttley. “We have built the infrastructure to become an auditorium. A few seats we’re filling in,” he said. “We’d like to build the auditorium with many seats.
He has added that another program, H1, is in the pipeline.
The upgrade also implies Honeywell’s machine is opened for businesses that can make their own unimaginably vast calculations. This product can cost about $10,000 an hour, Uttley says. While not telling how many clients Honeywell has, Honeywell has an agreement with JPMorgan Chase, whose quantum experts use their machine to perform gigantic duties, like building models for fraud detection. Intermediate Quantum Computing, Zapata Computing, and Cambridge Quantum Computing can provide queries in those enterprises without in-house Quantum Experts.
According to Uttley, with increased access to technology, quantum computers are getting close and used to fix climate change and pharmaceutical development issues. Uttley continues, according to Honeywell, the quantum volume of his machine will boost by ten each year during the next five years to 640,000, much higher than ever imagined.
In a quantum computer, the number of qubits is essential. More squabbles mean that scientists can resolve more complex issues with more possible solutions, with the possibility that organizations can be combined into a single virtual dispute allows much better computations. However, Qubit itself is a rough indicator of success, and therefore IBM has created a quantum quantity.
“IBM’s quantum volume benchmark is one of the earliest attempts to address this problem, and others will inevitably come down the pike as quantum computers become more capable,” said 451 Research analyst James Sanders. But for now, it’s easy to compare IBM and Honeywell machines at least.
Check out: IBM is quitting to develop facial recognition technology