INTELLIGENT TECHNOLOGY: QUANTUM COMPUTING

Breakthrough system that lets multiple quantum programs run on one machine at the same time.

Until now, even the smallest task required exclusive use of a quantum computer costing millions of dollars. Researchers often waited days in long queues while much of the hardware sat idle.

For cloud providers such as IBM, Google, and Amazon, HyperQ enables them to serve more users with existing hardware. For researchers, it means faster access to scarce resources.

Potential applications include drug discovery, where molecules could be modeled in hours instead of days, materials science, where lighter and stronger compounds could be designed faster and energy research, where smarter grids and advanced batteries could be developed on shorter timelines.

HyperQ solves this with“ quantum virtualization,” which creates isolated quantum virtual machines( qVMs) inside one device. Multiple users can then share the same machine without interfering with one another’ s work.

“ HyperQ brings cloud-style virtualization to quantum computing,” said Jason Nieh, professor of computer science at Columbia Engineering and co-director of the Software Systems Laboratory.“ It lets a single machine run multiple programs at once – no interference, no waiting in line.”

HyperQ acts as a quantum hypervisor – a software layer inspired by the virtualization that powers modern cloud servers. By dividing quantum hardware into smaller virtual machines and dynamically allocating jobs, it keeps machines busy and reduces delays.

Tests on IBM’ s largest quantum computers showed striking results. Wait times dropped by up to 40 times. Throughput increased tenfold. In some cases, performance improved as the scheduler directed workloads away from noisy regions of the chip.

“ Instead of forcing one person to monopolize the entire machine, many users can now share quantum resources at once,” Tao said.“ This changes the game for how quickly we can tackle some of the world’ s most challenging problems.”

HyperQ does not solve all of quantum computing’ s limitations. Qubits still have short lifetimes and high error rates, and large-scale error correction remains a major hurdle. But by breaking the one-program-ata-time bottleneck, Columbia’ s researchers say they have taken an important step toward making quantum computers more useful today.

The team is working to extend HyperQ across different architectures.“ The goal is to make sure it works not just on IBM’ s machines but on all types of quantum computers,” Nieh said. p

“ Previous efforts required specialized compilers and advance knowledge of which programs would run together,” said Runzhou Tao, lead author and now an assistant professor at the University of Maryland.“ Our approach adapts dynamically, which makes it practical for real-world use.”

28 INTELLIGENTCIO APAC www. intelligentcio. com