By Jerry Chow
This op-ed has been reprinted from IET Quantum Communication.
Classical computers just won't cut it when it comes to tackling the greatest challenges in biology and medicine. Even accurately simulating the relatively simple caffeine molecule, far simpler than DNA or even proteins, would take an impossibly large classical computer. That's why we need a paradigm shift in the field of computing, like a shift from the steam locomotive to a starship's warp drive. That's why we need quantum computers.
At IBM Quantum, we've already started building quantum computers based on superconducting transmon qubit technology, and have been making this transformative technology available to the world. Back in 2016, we made history when we put our quantum computers on the Cloud for anyone from enterprise developers to students to program. Last fall, we announced an aggressive roadmap toward building a 1121‐qubit device by 2023. However, even that device will not be large enough to solve science's greatest challenges. That will require fault‐tolerant quantum processors containing potentially millions or even billions of qubits working together.
The technology required to build a processor with a few million superconducting qubits will likely surpass the limits of dilution refrigerator scaling (Figure 1), so we have to come up with alternative ideas for scaling up our devices efficiently. We think that constructing the fault‐tolerant quantum computers of the future will require a distributed approach similar to today's supercomputers, where smaller processors each in their own fridge share quantum information coherently and without loss to tackle problems in parallel. This will only be possible if we can develop quantum interconnects to link processors together into a quantum intranet.
This requirement for a quantum intranet is distinct from the more commonly discussed context of a quantum Internet. Quantum communications advances typically center around data transfer security such as long‐distance entanglement and quantum key distribution. IBM has long been involved in this field; IBM researchers debuted the BB84 QKD protocol in 1984 [1], for example. However, we must not lose sight of the fact that short‐distance quantum interconnect technology may play a key role in extending computational systems to a scale sufficient for solving practical problems. A quantum Internet connecting and interfacing such scaled quantum computers would follow as a next step to enable new applications. [2].
Quantum interconnects capable of supporting our vision of a quantum intranet will require fundamental advances in physics. The no‐cloning theorem makes it impossible to convert an arbitrary, unknown quantum state into classical bits and then use these bits to recreate that exact state elsewhere, so we must devise schemes that stay coherent as we transfer quantum information from the states of superconducting qubits onto the states of photons, transmit these photons over wires, and recreate the quantum state on the superconducting qubits of another processor.
At the same time, the microwave photons we use to manipulate and read out our superconducting qubits require wiring overheads, and a significantly different cryogenic infrastructure to permit coherent transfer of quantum states even over short distances. Hence, converting microwave photons into optical photons allows for low‐loss quantum state transfer through fiber optic infrastructure that the telecom industry has spent decades maturing.
Academic groups have been working for over a decade to transfer quantum information from microwave to optical photons in a coherent manner with a variety of systems and mechanisms, and while this has not been conclusively demonstrated, progress has been made [3]. At IBM Research's Thomas J. Watson Research Center in Yorktown Heights, NY, we have been developing a silicon‐germanium microwave‐optical transducer (funded by the Army Research Office and the Laboratory for Physical Sciences) that's integrated with our own transmon qubit platform. Meanwhile, our IBM Research lab in Zurich is exploring a different platform, to develop piezo‐optomechanical transducers in gallium phosphide. Implementing and perfecting each of these solutions will be challenges on their own—plus, we must devise distributed error correcting codes to run on these systems.
 |
| IBM is developing a suite of scalable and increasingly larger and better processors, with a 1000‐plus qubit device targeted for the end of 2023. In order to house even more advanced processors to the thousands and eventually million‐plus qubit systems of the future, the IBM Quantum team is building a dilution refrigerator larger than any currently available commercially. (Credit: Connie Zhou for IBM) |
Our primary research focus is in scaling up transmon qubit
processors, but the issues surrounding scale‐up are important to every
qubit architecture. We think that quantum interconnect research, and a
quantum intranet, will be the likely solution to scale‐up across the
field. The community has already deemed quantum interconnects as
"crucial for sustained development of a national quantum science and
technology program" [4].
Other recent work involving connecting spatially separated qubits with
microwave includes Zhong, Y. et al.’s “Deterministic multi‐qubit
entanglement in a quantum network,” [5],
and Magnard, P. et al.’s “Microwave Quantum Link between
Superconducting Circuits Housed in Spatially Separated Cryogenic
Systems.” [6]
Given the importance of these devices, and the impact of scale, we cannot tackle these challenges alone. In August of last year, the US Department of Energy announced that it would allocate up to $625 million in funding over five years to support multidisciplinary Quantum Information Science (QIS) Research Centers. We expect that the work carried out at these centers will be essential to realizing the quantum intranet. Specifically, IBM Quantum will work with the Q‐NEXT center led by Argonne National Laboratory to co‐design technological building blocks and system needs that will allow us to realize these quantum interconnects.
The IBM Quantum team is developing quantum computers because we think that these devices have the capability to change the world, but seeing them through to completion will take innovative solutions to physics problems that physicists have never faced before. If we can realize quantum interconnects and a quantum intranet, then we think we have a clear path to making these devices, and their abilities, a reality.
References:
No comments:
Post a Comment