Toward the end of 2019, a flurry of announcements by some of the most prominent IT companies suggests that collaborations will become increasingly important in the quantum computing space as the players jockey for position in the nascent market. The companies in question include some of the biggest in the industry, including IBM and Microsoft, as well as Amazon Web Services, a new entrant in the field.
IBM has been especially adept at bringing international research groups into its Q Network, a cloud-based setup that provides access to the company’s quantum computing hardware, software, and developer tools to its selected partners. Initial research collaborators included Keio University (Japan), the University of Munich (Germany), the University of Melbourne (Australia), the University of Oxford (UK), the International Iberian Nanotechnology Laboratory (Portugal), the University of Montpellier (France), National Taiwan University (Taiwan), Oak Ridge National Laboratory (USA), and NC State University (USA).
In July of last year, the company announced it had gathered another set of European universities and research organizations into the Q Network. The most prominent institution in this latest clutch is the Swiss Federal institute of Technology in Zurich (ETH Zurich), but also includes Chalmers University of Technology in Sweden and Saarland University in Germany. The month previous IBM announced that the Consejo Superior de Investigaciones Científicas (CSIC) had established an IBM Q Hub in Spain. Together, researchers from these institutions will be involved in developing application software and system firmware for IBM quantum computers, while also providing training for engineering students at those institutions.
In September 2019, IBM teamed up with Fraunhofer-Gesellschaft, a leading institution for applied research in Germany, where it will have access to a Q System One machine, the company’s first cut at a commercial quantum computer. This 20-qubit system, which is installed at an IBM facility in Germany, will serve as a Q Network hub for European researchers and businesses allied with Fraunhofer.
Finally, in December IBM and the University of Tokyo jointly announced the Japan-IBM Quantum Partnership, a collaboration that will encompass other universities, as well as industry and government groups. As per the press release: “The partnership will have three tracks of engagement: one focused on the development of quantum applications with industry; another on quantum computing system technology development; and the third focused on advancing the state of quantum science and education.”
As in the Fraunhofer case, IBM will install a Q System One in Japan, the first in the region. Unlike the other research partnerships however, the University of Tokyo collaboration is slated to establish a quantum system technology center, which will be used to develop hardware and cryogenic and microwave technologies for future quantum computers.
Microsoft has come to the quantum computing space a bit more cautiously. The software-maker has been behind the curve in developing usable hardware, which is focused on a more exotic approach using Majorana zero modes-based topological quantum computation. Instead they have gotten into the game with their Q# offering, a language for developing quantum computing applications. More recently though, the Redmond gang developed a more complete quantum computing stack along with its newly hatched cloud-based Azure Quantum service and brought in a set of partners that offer workable quantum computers for that service.
Those hardware providers include IonQ and Honeywell, both of which have built quantum systems based on ion trap technology, as well as Quantum Computing Circuits (QCI), a Yale University spinout that has developed a novel way to error-correct superconducting qubits. Honeywell has been rather restrained with regard to the particulars of its hardware, while IonQ has been a lot more forthcoming. QCI has yet to publicly demonstrate its hardware, but its inclusion in Azure Quantum suggests a launch is imminent. Microsoft says it will also offer access to its own hardware based on topological qubits, but did not specify the timing of when that might happen. According to the company, the service will initially be offered in private preview in the coming months.
Not one to be left out of any paradigm-shifting computer technology, Amazon has thrown its hat into the quantum ring with a new AWS cloud offering called Amazon Braket. Announced last month in conjunction with a couple of other quantum initiatives, Braket is designed to be a “fully managed service that allows scientists, researchers, and developers to begin experimenting with computers from multiple quantum hardware providers.” Braket includes a Python-based software development kit (SDK) that developers can use to create quantum logic that can be tested in either a simulated environment or on the available hardware.
System providers include D-Wave, IonQ, and Rigetti. Together, they offer three distinct quantum computing approaches: quantum annealing (D-Wave), ion trap quantum computing (IonQ), and superconducting quantum circuitry (Rigetti). The AWS service will utilize the latest hardware from each of these providers, including D-Wave’s 2000Q machine and Rigetti’s 32Q device. Note that of all the big-name companies described here, Amazon is the only one (so far) that doesn’t offer its own proprietary quantum computing hardware.
In conjunction with the Braket launch, Amazon also revealed it has set up the AWS Center for Quantum Computing at Caltech that will take advantage of the local talent at the institute to “accelerate development of quantum computing hardware and software.” Also announced was the Amazon Solutions Lab, a program to hook up AWS customers with quantum computing experts both within and outside the company.
It’s no accident that all three of these providers are offering their quantum computing service via the cloud. Or maybe it’s more accurate to say that it’s no accident that all three of these companies are themselves cloud providers. The current crop of quantum computers are based on specialized hardware, most of which requires cryogenic refrigeration and a vibration-free environment. Even without the need to supply this kind of non-standard datacenter environment, the specialized nature of the devices and their small numbers lends itself more naturally to the abstractions of a cloud. It also makes it easier to add partners from far-flung locales, given that access to quantum machinery is just a browser click away.
Despite the different types of collaborations and collaborators we’ve described, the strategy of these three companies is roughly the same, namely to build the foundations of a new ecosystem, not so much to tie users to particular hardware – presumably, someday, quantum computers will be just another commodity – but to help support a toolchain and service framework that is unique to these IT providers. To that end, we expect all of them to continue to draw in additional research partners to help bolster their respective ecosystems.