-
Oracle made its first foray into the quantum realm with QMware
-
QMware offers a hybrid quantum computing service which combines quantum algorithms with advanced cloud computing to boost performance
-
Quantum computing is expected to be a multi-billion-dollar market by the end of the decade
Quantum computing isn’t quite ready for prime time, but Swiss startup QMware claims to have found a way to deliver the technology’s promised performance benefits today. And it’s teaming up with Oracle to bring that hybrid quantum compute power to the masses.
The partnership with QMware is notably Oracle’s first – and, for the time being, only – foray into the quantum computing space, an Oracle representative told Silverlinings. Other major cloud players, including AWS, Microsoft, Google Cloud and IBM, are already experimenting with quantum services.
Founded in 2020, QMware offers two primary products: quantum simulation software, which provides 42 logical qubits of processing power, and a hypervisor. For those who don’t know, a hypervisor is just software that allows you to run multiple virtual machines on a single computer.
As part of the partnership, QMware’s simulation software will run on Oracle Cloud Infrastructure powered by Nvidia A100 GPUs. The idea is to let QMware test and develop a hybrid quantum computing service that can then be made available to enterprise customers.
McKinsey & Co. tipped the financial services, pharmaceutical, chemical and automotive segments as the first industries to see a significant impact from quantum computing. But by 2030 and beyond, the telecom, oil and gas, sustainable energy, aerospace, electronics, semiconductor and logistics industries are also expected to get a quantum boost.
What might this look like in practice? QMware CEO Markus Pflitsch told Silverlinings that at a recent Oracle event in London, the pair showcased a quantum machine learning application which could be used to improve solar forecasting. The quantum model had an error rate of about 8%, which Pflitsch said is "about 41% better than other, non-quantum approaches." This kind of improvement could make it easier for utility companies to integrate renewable energy sources with historically volatile outputs (like solar) into the grid, he said.
All hail hybrid!
What QMware is doing with Oracle is known as hybrid quantum computing.
“There's a saying that there is no AI without the cloud – the same is true for quantum, in our perspective. There will be no commercially available quantum applications without cloud computing,” QMware CTO and co-founder George Gesek told Silverlinings.
Gesek said the idea that quantum computing will replace classical computers is a mistake.
“Quantum processors are here to speed up or enhance classical high-performance computers…in the future, we will see CPUs, GPUs and QPUs alongside [each other] in data centers,” the CTO stated.
Pflitsch added that by offering its simulation service via cloud providers like Oracle, it is aiming to help lower barriers to entry in the quantum realm and allow enterprises to develop and deploy new use cases like the one mentioned above. They can also reap performance benefits the simulator can offer in processing today.
“Yes, the future is disruptive with quantum, but you can already today beyond just dealing with toy problems unlock performance” using hybrid quantum computing, Pflitsch said.
How it works
Gesek and Pflitsch claimed QMware’s qubit simulation software not only performs better than current high-performance compute systems alone, but also today’s quantum chips. Qubits are just the basic units of information in the quantum world, akin to the bits used in regular computing.
But…how? (Yep, we wondered too.)
The CTO said it’s all about error rates. The simulator is able to complete processing using quantum algorithms (from Terra Quantum) with perfect simulated qubits, whereas native qubit systems are still error-ridden and low fidelity, Gesek said. Thus, it solves problems faster than a classical computer and performs better than still-imperfect qubit-based computers.
Emerging market
McKinsey & Co. predicted the quantum computing market will be worth $9 billion and $93 billion by 2040. IDC pegged the market at $1.1 billion in 2022 and tipped it to rise to $7.6 billion by 2027.
As of December 2022, McKinsey said $5.4 billion had already been invested and there were 223 start-ups operating in this space, according to the firm. The U.S. had the largest number of quantum computing startups, followed by Canada, the U.K., Japan, France and Germany.
While the technology is still in the process of maturing, McKinsey noted that quantum computing can already help speed up the time needed to solve simulation, optimization and cryptography problems as well as machine learning and artificial intelligence training.
“In the near term, leveraging a hybrid operating system to distribute a complex problem between [high-performance computing] HPC and [quantum computing] QC can bring a bigger computational advantage than either system alone,” the firm wrote.
Gesek said QMware is developing a new hybrid quantum computing tool capable of migrating existing algorithms to quantum algorithms to help accelerate existing IT infrastructure. It plans to release this “by the end of the year,” he said.