The Summit supercomputer in Livermore, CA, can crunch one thousand million million floating point operations 200 times a second. It’s held the medal for the most powerful computer in the world since 2018…but a field called quantum computing is coming for it.
To understand where we’re going, let’s start with where we are: The device you are using to read this encodes information in bits—aka binary digits—as 0s or 1s. A quantum computer uses qubits—aka quantum bits.
- Qubits can store information as 0 and 1 at the same time thanks to something called superposition.
- Qubits also exhibit a phenomenon called entanglement, in which each qubit, closely linked with other qubits, cannot be understood in isolation.
There are some computational problems so big that even the best supercomputer in the world can’t crunch the numbers. But a universal, fault-tolerant (more on that later) quantum computer that uses superposition and entanglement will be able to get the job done. When quantum computers can outperform classical computers (what we use today), we have reached “quantum advantage.”
Coming to a lab near you?
Researchers aren’t sure how long it will take to reach quantum advantage, but they can agree on one thing: Building a quantum computer is really freaking hard.
Encoding qubits requires you to control them at the atomic level. Some of the most advanced companies and labs working on this have quantum computers in the 50-100 qubit range, but experts estimate that a universal, scalable quantum computer will need hundreds or thousands of times that.
- Qubits destabilize easily. Quantum computers need to be error-corrected (protected from outside noise) and fault tolerant (capable of operating with an error rate below a certain threshold). Trust us, you can’t do this at home.
Nevertheless, quantum computing fundamentals are quite sound, according to Chris Monroe and Andrew Childs, two University of Maryland professors working in quantum and computer science.
In the next few years, small-scale experiments will overcome more engineering hurdles, but a working quantum computer with broad impact is still at least a decade away, Travis Humble, director of Oak Ridge National Laboratory’s Quantum Computing Institute, told the Brew.
Who’s playing: Outside of academia and research labs, Big Tech companies including Microsoft, Intel, IBM, and Google are working on quantum computers and accompanying suites of technology (cloud-based services, software). A handful of startups, including Monroe’s IonQ and Rigetti Computing, have also cemented their place at the forefront of the field.
- With expensive R&D but few near-term prospects for profit, it’s a hard field to break into without a lot of upfront capital. Big Tech has that part covered, but public companies still have to stop for check-ups every quarter-mile with shareholders (and quantum computing is an ultramarathon).
- Many experts have criticized some tech companies for skewing the narrative around quantum computing by stirring up the news cycle every few months boasting about what are actually incremental gains.
Don’t expect an overnight revolution
After the first computers were built, it took decades until the average business (let alone consumer) could have their own.
When quantum advantage is finally reached, it won’t flip a switch and flood the world with universal, scalable quantum computers, Jerry Chow, manager of IBM’s experimental quantum computing, told the Brew. It is an important benchmark for sure, but a train of incremental improvements and breakthroughs will need to follow to make quantum computers widely available and accessible.
Quantum computing will enhance and empower existing technology, potentially opening up capabilities like secure data teleportation between mobile devices, said Humble. But don’t expect to walk into Best Buy and pick up a quantum-enabled smart watch.
- Quantum computers will likely remain special purpose devices that interface with existing, classical computers. It will start with just a handful, and the cloud will serve as the point of access.
- Quantum computers won’t be used for all computation. Rather, workloads will be divided, with the bulk carried out on a classical computer and the parts that can’t otherwise be solved taking place on a quantum processor, Chow said.
Even if it doesn’t reinvent personal computing, quantum computing will raise the threshold of what is possible in other fields, and those effects will trickle down to average consumers.
Those opportunities include simulations and modeling in quantum mechanics, high energy physics, chemistry, and material science. That all sounds pretty conceptual, but it has huge implications for industries like pharma (through improved drug design) or energy (with breakthroughs in solid state batteries).
Until then…quantum computing needs time, money, and great minds.
If you’re interested, you don’t need to be a quantum physics Ph.D to help out. The field needs engineers to build circuits as well as computer scientists and developers to build out the software that will sit on top, according to Monroe.
On the business side, companies need to understand what value quantum computing can bring. Like machine learning, which is only at the tip of the implementation iceberg, it will require a fundamentally different way of looking at data and solving problems.
“We need the collective community to become more quantum literate,” to understand what problems they might be able to solve with it, Chow said.
We just threw a lot at you, so here are the main points
The promise: Quantum computers bypass the limits of binary systems and, one day, they’ll make supercomputers look like an abacus.
The roadblocks: Quantum computers today have 50-100 qubits…We need hundreds of thousands, but it’s hard to stabilize qubits and keep outside noise from interfering.
The projected timeline: Don’t listen to headlines: It’ll be years (or decades) before we have a scalable, fault-tolerant quantum computer.
The major players: Big Tech (Google, Microsoft, IBM, Intel), Startups (Rigetti, IonQ, D-Wave), Governments (China, U.S., Canada, EU)
This is a syndicated post. Read the original post at Source link .