“Is quantum computing real, or is this just more techno-hype?”
That’s the question I hear, and ask myself, quite often.
If you look at the market forecasts and the billions invested in quantum computing, you have to believe it’s real. Although that could be part of the hype as well.
For example, one analysis projected quantum computing to reach a value of USD 8.2 Billion by 2030 at a CAGR of 22% between 2023 and 2030 That’s just one of many that claim supersonic growth for quantum computing. We’re still waiting to see vendor revenue numbers and growth to support that estimate, by the way.
“But will it solve my problem? Someday?”
That’s a complex answer as of today. I believe that the truth is that it depends on the problem and the timing. And a whole lot more.
Quantum Computing Isn’t Hype, It’s a Major Innovation
Quantum computing is revolutionary technology. Of course we have a lot to learn and understand about it. That’s why we call it innovation.
Think back to the advent of the internet, which is a low-tech revolution compared to quantum innovation. Did we imagine when we used our first browser we would have today’s web? I doubt it.
That’s how tech innovation works. We innovate and get excited about the possibilities. We explore the potentialities as we spread the word of our incremental discoveries. Which often seem to be hypey when compared to in-place technology solutions.
Then comes the wait for the big bang… while we learn, we fail, we tune and learn again, apply what we’ve learned and with each cycle move the innovation forward. Again and again.
Remember your first word processor? Mine was a VI screen.
Your first cell phone? Mine came in a suitcase.
Your first personal computer? Whoo hoo, I had one 5 1/4″ floppy drive.
Yep, Evolution is part and parcel of innovation.
Quantum computing holds the potential for amazing value – but I believe we need to set some better expectations for non-quantum savvy folks.
That’s why the hype around quantum computing can be dangerous. The depth and breadth of quantum innovation means that many people don’t understand enough yet to separate the quantum hype from the quantum truth.
Small proof of concepts are touted as if they were true production results, coolness is announced with fanfare pretty much all of the time, and hope springs eternal. Big, boisterous hope for world changing impacts.
We get caught up in the coolness factor as we push toward the real grand slam. When will quantum computing solve some or all of the real-world computational limitations that organizations face today.
Such is the world of a new innovations. Quantum is no different, although I have to say that in my experience it’s one of the biggest media and investment extravaganzas, ever.
Fundamental Quantum Expectations
Part of the challenge is that we simply do not know, for 100% sure, that quantum computers will do what we expect them to do. We need more time and evolution before we can fully identify the real value of quantum. That takes innovation to mature early technologies, and innovation takes time.
So what is the reality for quantum computing, right now?
That depends on who you talk to. Let’s face it, different vendors and people have different perspectives, different truths if you will. That’s human nature. We all perceive the world uniquely.
First, let’s talk about what quantum computers are expected to do, and not do.
Quantum computers apply quantum mechanics to solve complex problems. They simulate large, complex computations and analysis in multi-dimensional spaces.
Translation? They can simulate real-world scenarios, and the impacts of changing variables in those scenarios, much faster and with more accuracy than any traditional computer.
Quantum computers are expected to accelerate complex computational applications such as machine learning and AI, simulations of complex problems in fields including chemistry, neuroscience, medicine and economics, and solve complex optimization problems (e.g., the traveling-salesman problem) in ways that traditional/classical computers simply can’t achieve.
Quantum computers will not replace traditional systems. They do not process transactions (they don’t have a database,) or run spreadsheets, word processors, websites or CRMs, to name a few. They simulate complex scenarios and solve complex computations. Period. So anyone who talks about quantum replacing traditional computing is creating hype that isn’t possible.
The most likely scenario for the majority of quantum computer use is as an “assist” to traditional computers. Meaning a blended or hybrid solution that uses quantum innovation to “power up” traditional systems to be far more effective in their work.
Standalone quantum systems will most likely appear for experiments and for very high-end computations, but I don’t see that being the primary implementation in the next decade. Why? Because they process simulations and computations, not the applications that drive businesses operations. Our traditional business applications and workflows will reach out to quantum computers to better perform their tasks. They will not be replaced by them.
Key Quantum Questions
I find that one of the best ways to learn is to ask questions. And there are so many questions about quantum to be asked. Following are the ones I ask myself and my vendor clients. I hear these same questions from enterprise and government users. Not the PhDs in math or physics, I’m talking about non-scientific humans.
Why don’t we know exactly what quantum computers do well? Where are the benchmarks? Supercomputers, big mainframes, OLTP’s TPC and more taught us the power of benchmarks. With quantum, the answer is we don’t know…yet.
That’s because we haven’t built a quantum computer that scales to run the problems that we expect them to solve. Quantum computers must import their data into the actual qubits within the system. They do not use databases to move data back and forth. That means that to solve a problem with 500,000 data points, we need a quantum computer able to import and hold 500,000 individual data elements. Today, there’s disagreement as to exact scale wrt variables, but I’ve seen ~4000 data elements processed effectively. We have a way to go when it comes to quantum innovation and scale – but we are learning every day.
Today we’re running “model” problems that have been mathematically created to resemble the full production problem – at a dramatically reduced scale. We are learning what quantum computers do, how they deliver results and examining those results for the value they deliver. As we scale these machines, we will create real benchmarks that represent full production value.
Why aren’t there software programs for quantum computers? Developing quantum programs is not the same as traditional programming. At all. We have to build highly complex quantum programs to run the computers; programs demanding specific quantum innovation that only highly skilled, hard to find and very expensive quantum experts can create. That’s a huge challenge that is often lost in the discussions and focuses on the quantum processing unit (QPU) hardware itself.
When will I be able to solve my problems on a quantum computer? The truth is, we don’t know. It depends on the size and complexity of the problem (e.g., the volume of data that has to be processed, the complexity of the problem) whether you are creating your own application, and how fast QPU vendors can scale their systems. Which leads to the next question….
With all the quantum hype, will we ever reach true quantum advantage? That depends on the problem. But what’s important is that everyone understands what reaching quantum advantage really means. As we all know, quantum advantage is defined as running a real-world problem on a quantum computer with better performance than a traditional computer.
We will reach that stage as QPUs expand their ability to solve large, production problems. We are not there yet. The problems used to demonstrate “quantum advantage” are select, very small problems that don’t attempt to reach the scale nor complexity of true production problems.
We all need to also understand that when we do reach quantum advantage, it will not be with a better price/performance than traditional systems. Most likely, it will come with an extremely high price/performance ratio. Once we reach quantum advantage, we can and will begin to focus on improving price/performance as the technology moves into more and more production instances.
What’s so hard about scaling these systems? There are a number of challenges in hardware design, but here’s one simple answer. As you scale QPUs, the inherent errors in today’s Noisy Intermediate Scale Quantum computers (NISQ) increases, geometrically. A microwave across the street has the potential to disrupt processing and collapse the quantum space. That means we need powerful error correction capabilities to reach the scale required for “production” problems. Large investments are being made to define and productize error correction and scale techniques, but we are not there yet, and we won’t be in the near term.
How do I get ready for quantum innovation? Do I even need to do anything yet? One of the misrepresentations that seems to have crept into the quantum computing hype is the idea that it’s all or nothing. Quantum computing has to be ready or we can’t benefit from it. That’s not the case. There are many things we can do today to prepare for quantum. Things we need to do to be ready for quantum computing when it matures. If nothing else, you must begin to establish a quantum workforce, whose training may take years.
Today, you can apply quantum methods to traditional computations and reap significant benefits. In fact, companies and vendors are applying quantum methods to solve quantum-possible problems on traditional machines using quantum simulators and new quantum algorithms. We are already moving forward at a rapid rate. The opportunity to learn and train is here. Either we begin to explore and learn, or we get left behind by competitors. It really is that simple.
The Bottom Line
Given the coolness factor of quantum computing, hype is expected. Just like every other tech innovation.
The difference is that quantum computing is one of the biggest innovation any in high technology have seen in our lifetimes. I believe it has the potential to be more disruptive than AI. The demand of AI for scale that it seems only quantum computing can deliver is going to be an interesting innovation journey.
I believe quantum computing will change our lives and “lnowns” when it comes to complex simulations and computations in specific arenas.
That said, quantum computing is not ready for primetime today. And it won’t be ready in a year, or three. There are too many barriers yet for quantum to provide scale, error correction, infrastructure and software needed for a production environment.
As with all advanced innovation, we must expect that these barriers will be removed over time. The only question is when we will be able to take full advantage of it. I expect that to become more clear over time, but don’t expect to run a major optimization problem on a quantum computer in the next 5 years. I suspect you’ll be holding your breath for a very long time.