Software to unlock quantum bottlenecks Japan's RIKEN Center Improves Quantum Algorithms
Of all the common sayings in computing, the phrase "if only software could catch up with hardware" probably ranks high, according to Tom's Hardware. However, software sometimes does catch up with hardware. In fact, for once, it appears that software can unlock quantum computing for classical computers. According to researchers at the RIKEN Quantum Computing Center in Japan, they have published a paper on an algorithm that can significantly accelerate specific quantum computing workloads. What's more, the workload itself - called the time-evolution operator - has applications in both condensed matter physics and quantum chemistry, two fields that could open up new worlds within our own.
Often, the improved algorithms will not be completely different; after all, updates are everywhere. Every application update, software update, or firmware upgrade is essentially bringing in revised code that solves a problem or improves performance (hopefully). Improved algorithms are great, as anyone with an AMD or Nvidia graphics card can attest. But let's face it: we're used to being disappointed by performance updates.
Yet in this case, the performance improvement is extraordinary. Indeed, the results are impressive. With improved algorithms (themselves a hybrid of quantum and classical methods), future quantum computers can become simpler than we thought: they will be able to solve larger problems faster than we expected, and at a lower cost. But the performance gains don't stop there. They could enable conventional machines to handle levels of complexity that only quantum computers can solve.
"Time-evolution operators are huge lattices of numbers that describe the complex behavior of quantum materials," explains Kaoru Mizuta of the RIKEN Center for Quantum Computing. "They are very important because they provide very practical applications for quantum computers - to better understand quantum chemistry and solid state physics."
The algorithm improvements eliminate the Trotterization technology quantum computer deployed so far - one that has been suspected of not scaling in the long term. That's because the technology requires a large number of quantum gates, each requiring a variable number of quantum bits to perform a given function. Even IBM's 1121 quantum bit-count Condor QPU (Quantum Processing Unit), which will be released this year, will have difficulty enabling the quantum gates that Trotterization is expected to require to handle workloads that actually make sense in quantum computing.
No, quantum computing will not be in our smartphones. In a way, today's superconducting coolers are comparable to ENIACs before the advent of integrated microchips. Or from that point to the equivalent of today's fastest CPUs or best GPUs. that's the quantum path ahead of us - the one that the opening line just sounded.
