To appreciate the significance of Quantinuum’s latest research, it is important to first understand why quantum error correction plays such an important role in quantum computing.
Solutions to world-class problems such as climate change, new pharmaceuticals, custom design of new materials, long-range electric vehicle batteries, and many other applications are beyond the computing power of today’s most powerful supercomputers.
A quantum computer is not just a faster or bigger type of computer. It’s essentially a different type of computing technology, anchored in the weird and wacky world of quantum mechanics. A quantum computer has the potential to solve huge and complex problems quickly, but only if it is equipped with the huge number of qubits (quantum bits) necessary to do the job.
For example, a classical computer would probably never be able to crack the Bitcoin encryption key, even if given the remaining lifetime of the universe to solve it. According to University of Sussex in the UK, it would require a quantum computer with 13 million qubits running for about 24 hours to crack Bitcoin’s key. Pumping the qubit count to 300 million qubits would reduce the quantum computer’s solution time to about an hour or less.
For perspective, today’s quantum computers have small qubit counts, ranging from 50 to several hundred qubits with the potential to have several thousand qubits in a few years. As the University of Sussex example shows, this is still a small fraction of the number needed to do serious and useful calculations.
Can’t we just add lots of qubits to a quantum computer?
Physics and engineering considerations related to qubit fidelity and error correction limit the ad hoc addition of large numbers of qubits to a quantum computer. Quantum scientists have yet to develop a usable and scalable method for error correction.
Classical computers rarely make mistakes, so it doesn’t make much difference if a few bits are flipped by trillions of calculations. Unlike classical computer bits that operate strictly as 1’s or 0’s, qubits operate in quantum superposition states without the precision of being exactly 1’s or 0’s.
Qubits are also very sensitive to errors caused by environmental factors such as noise, wiring, and even other qubits. Qubit errors can occur even when exposed to relatively weak galactic cosmic radiation. Additionally, a qubit’s quantum state decays rapidly, requiring a quantum computer to start and complete its entire set of operations before the quantum states collapse. It is no exaggeration to say that every part of the quantum computing process is a potential source of qubit errors.
Quantum error correction (QEC) is complicated not only because of its quantum nature, but also because there are multiple types of qubit errors. Depending on the quantum technology and process, error counts can range from one error per hundred calculations to one error per many thousands of calculations.
Error correction is essential because it will allow us to build large, fault-tolerant quantum computers, scalable to hundreds of thousands of error-corrected qubits.
Significance of Quantinuum’s fault tolerance achievement
Quantinuum published the first research paper to demonstrate an end-to-end fault-tolerant circuit with entangled logic qubits using real-time error correction. It is also the first time that two error-corrected logical qubits have performed a circuit with higher fidelity than its physical qubit components.
What’s important is that Quantinuum’s demonstration of fault tolerance creates a new starting point that could allow future researchers to expand the number of qubits.
It is important to note that Quantinuum’s QCCD architecture has greatly contributed to the company’s ongoing research and enabled experimentation with geometries. The flexibility of QCCD bands allows qubits to be rearranged arbitrarily and experimentally to accommodate codes with exotic geometries and codes that are not in one or two dimensions, particularly compared to what is possible with quantum computers that have fixed geometries. The QCCD design was first proposed by David Wineland’s group at NIST in a 1998 document.
Although the original QCCD architecture contained some unresolved technical issues, Tony Uttley, previously president of Honeywell Quantum Systems, and the Honeywell team decided to develop the company’s next-generation quantum system using the QCCD architecture. Fully aware of the risks, Uttley decided that the opportunity for greater rewards outweighed what he believed to be manageable risks.
Considering Quantinuum’s technical achievements in 2022 and earlier, the decision to use the QCCD architecture has proven correct.
Connect the dots 2022
The following list details the advances made by Quantinuum that have established foundational work for further research in 2022 and beyond.
- March 3 – A world record SPAM (Status Preparation and Measurement Error) using barium-137 provided measured indications of a near-term future with SPAM error rates at 105 range. Improving SPAM fidelity helps reduce the errors that accumulate in today’s “noisy” quantum machines, which is critical for the transition to fault-tolerant systems that prevent errors from cascading through a system and destroying circuits.
- April 14 – Quantinuum’s sixth quantum volume record was measured at 4,096 (212). The QCCD architecture allowed an increase in qubits and a corresponding increase in fidelity. Important because the increase in fidelity is necessary when more qubits are added along the way to ensure that it will be computationally useful. The Quantinuum H1-2 system used all 12 of its qubits for this new milestone which signals the possibility that Quantinuum will soon add more qubits.
- May 24 – InQuanto released. It is a quantum computational chemistry software platform built for computational chemists. This platform could only be performed accurately using a high-performance quantum hardware system such as Quantinuum’s H1 series.
- June 14 – Upgrade from 12 to 20 qubits on the H1-1 machine. The number of gate zones in the QCCD architecture was also increased from 3 to 5 to allow for more concurrent operations and improved parallelism in circuit execution. Previous planning and work in 2021 set the stage for this upgrade.
- July 11 – Barium and ytterbium junction transport, a measured scaling method exploiting 2-D lattices. It allows two different ion species to move through a compound in a surface trap together as a pair. This will be incorporated into the future design of the System Model H3. It is expected to help scale and provide faster computations, allow more qubits to be added and reduce errors.
- July 20 – New phase of matter realized in H1-1 as described in the research paper: “Dynamic topological phase realized in a trapped quantum simulator” (peer review published in Nature for the 2021 paper)
- August 4 – New quantum dynamics simulation technique demonstrated as described in the research paper: “Holographic dynamics simulations with a trapped ion quantum computer” (peer reviewed published in Nature Physics for the 2021 paper)
- August 4 – This article is based on this research paper: “Implementing Gate Jam Fault Tolerance in Five-Qubit Code and Color Code”. This paper documents a future where real-time quantum error correction paves the way for a fault-tolerant regime.
The bottom line
Real-time error correction is essential for the continued development of reliable large-scale quantum computers. Error correction is a high priority for almost every company in the quantum ecosystem. This is why a lot of research is being done by various companies and universities.
Quantinuum has taken a small but very important two-qubit step towards fault tolerance. It opened the door to a new and promising direction of research.
Without fault tolerance, today’s quantum computing technology will not be capable of solving the important world-class computing problems we hoped it could solve. So the question is – can we do it? In my opinion, absolutely.
Note: Moor Insights & Strategy writers and editors may have contributed to this article.
Moor Insights & Strategy, like all research and technology industry analyst firms, provides or has provided paid services to technology companies. These services include research, analysis, consulting, consulting, benchmarking, acquisition matching and speaking sponsorships. The company has had or currently has paid business relationships with 8×8, Accenture, A10 Networks, Advanced Micro Devices, Amazon, Amazon Web Services, Ambient Scientific, Anuta Networks, Applied Brain Research, Applied Micro, Apstra, Arm, Aruba Networks (now HPE), Atom Computing, AT&T, Aura, Automation Anywhere, AWS, A-10 Strategies, Bitfusion, Blaize, Box, Broadcom, C3.AI, Calix, Campfire, Cisco Systems, Clear Software, Cloudera, Clumio, Cognitive Systems , CompuCom, Cradlepoint, CyberArk, Dell, Dell EMC, Dell Technologies, Diablo Technologies, Dialogue Group, Digital Optics, Dreamium Labs, D-Wave, Echelon, Ericsson, Extreme Networks, Five9, Flex, Foundries.io, Foxconn, Frame ( now VMware), Fujitsu, Gen Z Consortium, Glue Networks, GlobalFoundries, Revolve (now Google), Google Cloud, Graphcore, Groq, Hiregenics, Hotwire Global, HP Inc., Hewlett Packard Enterprise, Honeywell, Huawei Technologies, IBM, Infinidat, Infosys , Inseego, IonQ, IonVR, Inseego, Infosys, Infiot, Inte l, Interdigit al, Jabil Circuit, Keysight, Konica Minolta, Lattice Semiconductor, Lenovo, Linux Foundation, Lightbits Labs, LogicMonitor, Luminar, MapBox, Marvell Technology, Mavenir, Marseille Inc, Mayfair Equity, Meraki (Cisco), Merck KGaA, Mesophhere , Micron , Microsoft, MiTEL, Mojo Networks, MongoDB, MulteFire Alliance, National Instruments, Neat, NetApp, Nightwatch, NOKIA (Alcatel-Lucent), Nortek, Novumind, NVIDIA, Nutanix, Nuvia (now Qualcomm), onsemi, ONUG, OpenS , OpenS, Oracle, Palo Alto Networks, Panasas, Peraso, Pexip, Pixelworks, Plume Design, PlusAI, Poly (formerly Plantronics), Portworx, Pure Storage, Qualcomm, Quantinuum, Rackspace, Rambus, Rayvolt E-Bikes, Red Hat, Renesas , Residio, Samsung Electronics, Samsung Semi, SAP, SAS, Scale Computing, Schneider Electric, SiFive, Silver Peak (now Aruba-HPE), SkyWorks, SONY Optical Storage, Splunk, Springpath (now Cisco), Spirent, Splunk, Sprint ( now T-Mobile), Stratus Technologies, Symantec, Synaptics, Syniverse, Synopsys, Tan ium, Telesign, TE Connectivity, TensTorrent, Tobii Technology, Teradata, T-Mobile, Treasure Data, Twitter, Unity Technologies, UiPath, Verizon Communications, VAST Data, Ventana Micro Systems, Vidyo, VMware, Wave Computing, Wellsmith, Xilinx, Zayo , Zebra, Zededa, Zendesk, Zoho, Zoom and Zscaler. Founder, CEO and Chief Analyst of Moor Insights & Strategy, Patrick Moorhead, is an investor in dMY Technology Group Inc. VI, Dreamium Labs, Groq, Luminar Technologies, MemryX and Movandi.