freshidea - Fotolia
While quantum computing is still a ways off from mainstream enterprise adoption, two trends around the technology are already clear: It will require users to completely rethink their computing infrastructure, and, in most cases, will be tightly intertwined with the cloud.
Quantum computing replaces traditional binary computing systems, where bits are either a 1 in an "on" state, or a 0 in an "off" state. It applies recent advances in physics and virtualization to increase the possible combinations of states by factors of 10 or more.
That change is expected to dramatically increase system processing power -- but by how much, exactly, remains unclear. On the low end, advocates claim that quantum algorithms will run 10 to 20 times faster than current computing systems, while some observers think the improvements will be by the hundreds.
Quantum computing to find a home in cloud
While it could provide a significant performance boost to certain applications, such as those that use AI and machine learning, quantum computing remains daunting to deploy. At its core, quantum computing is incompatible with all current computer infrastructures; to use it, organizations will need new servers, software, storage and networks.
"Because quantum uses different semantic constructs than classical computer systems, the industry will need to rethink and redesign how applications run," said Chirag Dekate, research director at Gartner.
Consequently, the technology is too complex for enterprises to deploy in their data center, which means they will largely consume cloud-based quantum computing services, said Peter Rutten, research manager at IDC.
Engineering teams in established tech organizations, such as Google, IBM and Microsoft, as well as in startup companies like D-Wave and Rigetti Computing, are attempting to commercialize quantum computing. They aim to create computing ecosystems, including programming and management tools, which can help enterprises build applications that take advantage of this extra processing power.
Some early cloud-based quantum computing services have started to take shape. IBM, for instance, has encouraged researchers to dabble with its offering for a few years. Google, for its part, developed Bristlecone, a quantum processor, and Rigetti, which has raised $119.5 million in venture capital funding and applied for more than 50 patents, has been soliciting beta testers for its cloud platform.
More theoretical than practical -- for now
But, even when they rely on cloud-based quantum computing, enterprises face a learning curve with the technology.
"Quantum computing is incredibly complicated," said Charles King, analyst at Pund-IT. "Most users will struggle just to understand what it is -- let alone try to deploy it."
The application and infrastructure development work around the technology is still in an early state. While it's much more than just words on the page of a technical journal, cloud-based quantum computing is still eons away from being something users can deploy with a few mouse clicks.
Quantum systems also rely on emerging technology, such as photon detectors, which are complex to build. As a result, the cost of quantum services is very high today. "In some cases, companies are paying $2,000 an hour for quantum services," Gartner's Dekate said.
The end result is a technology that, at least for now, is beyond the reach of most organizations.
"Quantum computing is in its infancy," King said. "Its appeal is limited to select vendors with deep research and development experience, universities [and] researchers and businesses with substantial research and development budgets."
Expect increased experimentation over the next year or two, with a few targeted commercial applications appearing in three to five years, he added.