A research team at the University of Science and Technology of China (USTC) has introduced Zuchongzhi-3, a 105-qubit superconducting quantum computer capable of performing calculations at speeds 10¹⁵ times faster than the most powerful supercomputer and one million times faster than Google’s latest quantum results.

Zuchongzhi-3 represents a significant step forward in quantum computing. The research team announcing the breakthrough was led by Jianwei Pan, with Xiaobo Zhu, Chengzhi Peng, and others, publishing their findings in Physical Review Letters.
Quantum supremacy refers to the point where a quantum computer can calculate faster than even the most powerful classical computers. In 2019, Google’s 53-qubit Sycamore processor completed a specific sampling action in 200 seconds, which is anticipated to take 10,000 years on a classical supercomputer, for example.

However, in 2023, USTC researchers revealed that an improved classical algorithm could solve the same task in 14 seconds using over 1,400 A100 GPUs. Using the Frontier supercomputer’s enhanced memory, the task can now be completed in 1.6 seconds, putting Google’s past claim of quantum supremacy into question.

Technical specifications and performance

Zuchongzhi-3, which builds on the 66-qubit Zuchongzhi-2, outperforms previous models in key areas. The system has a coherence time of 72 microseconds, with single-qubit gate fidelity at 99.90%, two-qubit gate fidelity at 99.62%, and readout fidelity at 99.13%. The improvements allow for more accurate and complex quantum operations.

The research team performed an 83-qubit, 32-layer random circuit sampling task at processing speeds 15 orders of magnitude faster than the most powerful classical supercomputer. Zuchongzhi-3 also outperformed Google’s latest quantum computing results by six orders of magnitude, setting a new benchmark for superconducting quantum systems.

What is quantum supremacy?

Quantum supremacy has broad implications across industries.

• In cryptography, quantum computing could run Shor’s algorithm, which threatens current encryption methods by efficiently factoring large numbers.
• In drug development, quantum simulations could accelerate the discovery of new treatments by modelling molecular interactions with high precision.
• Quantum-powered weather forecasting could improve the climate model accuracy, and quicker market analysis and algorithmic trading could improve stock trading efficiency.
• Quantum computers could help material scientists identify new materials with unique properties by performing precise atomic-level modelling.

Challenges and progress

Despite the promise of quantum computing, there are significant challenges to achieving consistent quantum supremacy. Qubit stability and error rates remain key concerns. Quantum operations are particularly sensitive to external noise, which raises the likelihood of computation errors.

The number of functional qubits is another limiting factor. Building scalable quantum systems requires increasing the number of qubits while keeping error rates low. Current developments in error correction, like surface code integration, try to address these concerns.

USTC researchers are working on improving error correction by starting with a distance-7 surface code and planning to extend this to distances 9 and 11 to improve fault tolerance and system stability.

Global race in quantum computing

Investment in quantum computing has grown rapidly, with major players like Google, Microsoft, and IBM driving research and development. Google remains focused on fault-tolerant quantum computing, while Microsoft is looking at topological qubits for scalable quantum systems.

IBM has made progress in superconducting quantum processors, whereas Alibaba has focused on quantum cryptography and simulation. China and the United States continue to lead the way in quantum research, with government and private sector funding driving advances in hardware, algorithms, and applications.

Collaborative effort

The research involved multiple institutions, including the Shanghai Research Centre for Quantum Sciences, Henan Key Laboratory of Quantum Information and Cryptography, China National Institute of Metrology, Jinan Institute of Quantum Technology, the School of Microelectronics at Xidian University, and the Institute of Theoretical Physics under the Chinese Academy of Sciences.

The next step for USTC researchers will be to scale the system while improving accuracy and reducing error rates – important steps toward achieving reliable quantum supremacy.

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.

The post Zuchongzhi-3 outperforms Google in quantum computing speed appeared first on TechWire Asia.

Its pervasive influence comes when the digital world faces a new raft of data challenges. According to Verhoef, humanity has generated and stored 175 zettabytes of data – a volume so massive that, if stored on CD-ROMs and stacked, it would create a tower reaching the sun and back.

This (literally) astronomical amount of data, comprising both structured (roughly defined as database entries) and unstructured (photos, videos) content, presents challenges and opportunities for infrastructure development.

“Even though there’s a lot of talk about tape being dead, the big hyperscalers are the biggest customers of IBM tape,” Verhoef said, stating cloud providers still rely on tape libraries for cost-effective storage of rarely-accessed data. This practical approach to data storage demonstrates how ‘traditional solutions’ continue to play a crucial role in modern infrastructure.

Breaking through computing bottlenecks

IBM’s infrastructure strategy focuses on overcoming three bottlenecks: software infrastructure, compute capacity, and storage access times. The company has made significant advances in all three areas, Verhoef said.

IBM’s acquisition of Red Hat and its OpenShift containerisation platform has changed application deployment. Traditional deployment cycles that once took up to four months can now be significantly shortened, allowing businesses to respond more rapidly to market demands.

On the compute front, IBM has pushed the boundaries of chip density, progressing from 7-nanometre chips to developing 2-nanometre technology. The advancement isn’t just about speed – it’s about reliability, too. NASA’s choice of IBM chips for the Mars rover demonstrates the technology’s proven track record in mission-critical applications where failure isn’t an option.

Storage technology has seen perhaps the most dramatic improvements. The evolution from traditional hard drives (10,000 microseconds access time) to flash storage (80 microseconds) and now to NVME-RDMA technology (7 microseconds) has transformed data access capabilities. IBM’s Flash System, incorporating these advances, is one of the leading technologies of storage.

The quantum computing revolution

Perhaps the most exciting development in IBM’s infrastructure journey is its quantum computing breakthrough. The company progressed from a 27-qubit system in 2019 to an impressive 433-qubit computer today. The quantum system operates at 50 millikelvin – colder than outer space – and has allowed a fundamental shift in computing paradigms.

“A quantum computer is not a new car or a supercar. It’s more equivalent to a boat that we place in the water, and we are exploring areas unknown,” Verhoef said, describing how quantum computing opens up entirely new possibilities rather than improving traditional computing’s capabilities.

Security implications and future challenges

The advent of quantum computing brings both opportunities and challenges. A 10,000-qubit computer could break current RSA encryption algorithms in less than five seconds, prompting IBM, among others, to develop quantum-safe encryption systems. It’s already used quantum-safe encryption in its financial systems, and Verhoef said the proactive approach to security demonstrates the company’s commitment to responsible technological advancement.

The path to cognitive computing

IBM’s journey in artificial intelligence predates current AI trends by decades. From Deep Blue’s historic chess victory over Garry Kasparov in 1997 to Watson’s Jeopardy triumph in 2011, IBM has pushed the boundaries of what we now call cognitive computing. The achievements weren’t just publicity stunts but significant milestones in natural language processing and decision-making capabilities.

Looking ahead: The next 50 years?

The implications of technology for businesses and society are profound. Verhoef noted that quantum computing could help solve problems that would otherwise require “25% of the earth’s resources” with traditional computing methods. The efficiency gain could revolutionise fields from drug discovery to climate modelling.

Current CPU technology has reached the processing equivalent of a mouse brain, but the journey toward human intelligence-level computing continues. As we approach what futurists call “singularity” – the point where computer processing power surpasses human cognitive capabilities – the role of infrastructure becomes increasingly important.

IBM’s progression from traditional infrastructure to quantum computing mirrors technological advancement – it has led to a reimagining of how we process, store, and use information. On the brink of a quantum revolution, the next chapter in computing infrastructure promises to unlock possibilities we’re only beginning to imagine, from solving complex molecular problems to revolutionising financial models and beyond.

The post IBM’s quantum leap: From traditional infrastructure to 433-qubit computing appeared first on TechWire Asia.