eWEEK content and product recommendations are editorially independent. We may make money when you click on links to our partners. Learn More.
Intel CEO Brian Krzanich came to this year’s Consumer Electronics Show with the future in mind even as his company and the rest of the IT industry continue to deal with the fallout from the vulnerabilities in processors that have given rise to the threats of Spectre and Meltdown.
During his keynote address at the giant tech show, Krzanich said Intel will be rolling out fixes to the security problems in its chips later this month, and that there doesn’t yet appear to have been a hack exploiting the vulnerabilities. However, much of the focus of his talk was on strides Intel is making in the areas of quantum and neuromorphic computing, emerging technologies that hold the promise of new generations of computers that will be able to run highly complex workloads significantly more quickly than modern supercomputers.
The CEO held up Intel’s latest quantum test chip, the 49-qubit “Tangle Lake” processor that brings Intel into greater competition with the likes of IBM—which reportedly has a 50-qubit chip that it brought to CES—Google and Microsoft, all of which are putting large amounts of resources and money into quantum computing initiatives. Tangle Lake also comes just about three months after Intel introduced a 17-qubit chip, with Krzanich noting the company’s rapid innovation in chips from 7 qubits to 17 to 49 (pictured).
In addition, he spoke about a neuromorphic research chip, code-named “Loihi,” that will be shared with universities and research institutions to use with increasingly complex artificial intelligence (AI) workloads. Loihi put training and inference onto a single chip, which Intel officials say will make machine learning more energy-efficient.
“This has been a major research effort by Intel, and today we have a fully functioning neuromorphic research chip,” Krzanich said. “This incredible technology adds to the breadth of AI solutions that Intel is developing.”
The push for new capabilities like quantum computing and neuromorphic processors—which are being designed to mimic the way the human brain gathers and processes data—is being accelerated with the rise of such emerging technologies as AI, machine learning, data analytics and virtual reality, and the rapid growth in the amount of structured and unstructured data being generated. It will enable organizations to address such problems as drug development, financial modeling and climate forecasting much faster than the months or years it takes current supercomputers.
Quantum computing promises systems that are multiple times faster than current supercomputers. Organizations for the past several years have been relying on faster CPUs, accelerators like GPUs and field-programmable gate arrays (FPGAs), and faster interconnects to accelerate their work. At the core of quantum computing are qubits. Current systems use bits that hold values of 0 or 1. But qubits—or quantum bits—can be 0 and 1 at the same time, opening the possibility of systems running through millions of calculations simultaneously and at high speeds.
However, there are challenges, including the fragility of the qubits themselves. While they can be entangled—sharing the same state with two or more qubits—they can revert back to one of the two states if affected by an outside source like noise, which can lead to data loss. To counter that, they need to operate in extreme cold environments—about 20 millikelvin, or 250 times colder than deep space—which means that packaging of the qubits is critical.
According to Intel officials, the name of the chip reflects such challenges: Tangle Lake is named after a chain of lakes in Alaska (extreme cold temperatures) and points to the fact that qubits need to be entangled to function.
Tangle Lake is another step in Intel’s plans to develop a complete quantum computing system—including the architecture, algorithms and control electronics—and hitting the 49-qubit mark will help researchers improve error correction techniques and simulate computational problems.
“In the quest to deliver a commercially viable quantum computing system, it’s anyone’s game,” Mike Mayberry, corporate vice president and managing director of Intel Labs, said in a statement. “We expect it will be five to seven years before the industry gets to tackling engineering-scale problems, and it will likely require 1 million or more qubits to achieve commercial relevance.”
Officials with D-Wave, in which Google has invested, say they have quantum computing systems on the market, though some in the industry have said they aren’t true quantum systems.
As part of Intel’s efforts, the company is researching different types of qubits. Tangle Lake involves superconducting qubits, but researchers also are working with spin qubits, which are much smaller than superconducting qubits. They look like a single electron transistor, according to company officials. The company has created a spin qubit fabrication flow on its 300mm process technology.
While quantum computing will address future workloads, neuromorphic chips are being designed to deal with applications that demand real-time processing of data, such as with smart security cameras and smart-city infrastructures.