Watch the IT industry long enough and you develop a healthy skepticism for anything labeled “new” or “unique” but now and then something thoroughly different than most anything that has gone before comes along. A few cases offer such clear opportunities for advancement that the product’s evolution becomes a subject of common interest to the point that it “grows up in public.” Quantum computing, and the recent IBM Q announcement are good examples of this.
Getting from classical to quantum
IBM’s press release noted the company’s decades of quantum computing research that resulted in quantum processor fabrication, system design, materials development and software and programming tools. Those efforts came together in May 2016 when the company launched the IBM Quantum Experience service which is available to anyone through IBM Cloud.
The Quantum Experience enables users to run algorithms and experiments on IBM’s quantum processor, work with individual quantum bits (qubits—quantum processors), and explore tutorials and simulations. In less than a year, 40,000+ users from over 100 countries have run more than 275,000 experiments on the IBM Quantum Experience. The service has also inspired fifteen third party research papers that have been posted to www.arXiv.org, Cornell University’s online research archive, and five of those were published in leading journals.
How does this relate to IBM’s new press release? Most importantly, the company released an initiative and roadmap for building “commercially available universal quantum computing systems” but what exactly does that mean? According to IBM, there are three forms of quantum computer:
- Quantum Annealers with 1 to 49 qubits that are mainly used for optimization problem experiments. Annealers’ performance is analogous to common “classical” computing systems;
- Approximate Quantum systems that range in size from 50-100 qubits are notably faster and more powerful than any classical systems. As a result, they are useful for quantum chemistry, materials science, sampling, quantum dynamics and optimization problem applications; and
- Universal Quantum computers capable of scaling to thousands of qubits, are orders of magnitude faster and more powerful than classical computers and can be applied across the same use cases as Approximate Quantum systems, as well as in secure computing, machine learning, cryptography, and advanced searching.
In essence, IBM believes it has found the means to scale the technologies developed for the Quantum Experience into saleable, commercial grade Universal Quantum systems, and has set a workable roadmap in place. Given the potential value quantum computing offers in a wide range of technical and scientific areas, that is a very big deal.
Taking the quantum leap
Should we believe IBM? Healthy skepticism is wise whenever you’re dealing with new IT developments so it’s worth asking why the company would make such a claim unless it believes it has a way forward. Truth be told, lying about so high-profile a subject would seriously damage IBM’s credibility with customers, partners and research communities. Those are long term risks that would far outweigh any short-term gains.
So let’s take IBM at its word about a way forward to commercial universal quantum systems. The announcement also noted other sizable steps leading toward the company’s achieving its goals:
- A new API that developers can use to build interfaces between the five qubit Quantum Experience system and classical computers, and
- An upgraded simulator for the Quantum Experience capable of modeling circuits with up to 20 qubits, along with a full software developer kit (SDK) for building simple quantum applications and programs.
IBM also detailed engagements with researchers and academic institutions, including MIT, the University of Waterloo and the European Physical Society to leverage the IBM Quantum Experience for educational purposes. In addition, the company described work it is doing with Samsung, JSR, Honda, Canon, Hitachi Metals and Nagase through the IBM Frontiers Institute which develops and shares groundbreaking technologies, including quantum applications, and evaluates their business and industrial implications.
In other words, along with actively developing commercial Universal Quantum computers, IBM is also proactively building the markets and demand for those devices. This may seem simply mercenary to some people but it illustrates how practical value is key to the development of new technologies. Few, if any, IT vendors engage in pure research.
The IBM Q effort also provides a great example of how vendors bring essentially unique products to market. Since quantum computers are hugely different in form and function from classical systems, commercializing them requires sizable education efforts by vendors, universities, businesses and industries. It’s difficult or impossible for people to imagine, let alone build something entirely new unless they understand the tools that are available to them.
Why IBM Q?
What drives IBM’s commercial interest in the quantum computing? For one thing, the massive improvements in parallel performance offered by quantum systems are staggering. It’s been estimated that the performance of a modest 30 qubit quantum system could equal a classical computer running at about 10 teraflops. That would nearly place it among the top ten systems on the most recent (November 2016) Top500.org supercomputer list.
Quantum computers can also tackle problems that are too complex and exponential for classical systems. Dr. Jerry M. Chow, a manager and staff member in Experimental Quantum Computing at IBM’s T.J. Watson Research Center who recently briefed analysts on the IBM announcement noted that while “Classical computers are really good at crunching through known data, quantum computers excel at working with and exploring experiential data.”
How so? Quantum systems could vastly speed computational processes and analyses related to discovering new drugs and materials for industry. Quantum capabilities could also be used to develop cryptographic algorithms and codes that are unbreakable with conventional systems. Supply chain processes, financial services and artificial intelligence are other areas where quantum breakthroughs could result in game-changing developments.
In other words, it’s no wonder that IBM is putting its substantial investments and energies behind IBM Q.
Okay, so how soon will it be before IBM Q starts delivering on its substantial promise? Despite the progress that the company has made to date, quantum solutions and their market opportunities are in early days. The again, virtually every transformational technology—time and navigation instruments, internal combustion engines, power generation and delivery, railway and road construction, computer hardware and software—required significant research and investment to achieve their revolutionary potential.
IBM isn’t the only vendor developing quantum computing technologies but this recent announcement suggests that the company is gaining significant traction and momentum that will eventually result in commercially viable IBM Q systems and markets ready for those solutions.