Quantum Computing in the Roadmap to Greener Calculations
Alexandra De Castro
October 5, 2023
The impact of climate change is increasing in intensity and frequency. Since the last decade, we have experienced more and more extreme weather events than in the past. Wildfires are more common, even in unexpected places and the steady rise of the oceans will keep dramatically deteriorating half of the world's population's quality of life.
Scientists have been using the most powerful supercomputers to store and process weather and climate-related data, simulating patterns and modeling the future implications of climate change. Ironically, these data centers consume a huge amount of energy!
Since the last century, high-performance computers have been growing, offering more precise and faster calculations, and today, they accommodate thousands of processor cores requiring entire buildings with costly cooling systems. For this reason, rethinking energy consumption and reducing green house emissions has become an important mission for enterprises offering computational and data storage services. However, despite the efforts to lower energy consumption, they are still extremely high. Let's see some numbers.
Classical supercomputers’ footprint
Data centers' consumption accounted for around 1.7% of the global electricity demand during 2022. A data center can consume the same amount of electricity as a town. If you think this is an exaggeration, let's take, for instance, the Frontier supercomputer, fabricated by Hewlett Packard and hosted at the DoE Oak Ridge Laboratory in Tennessee, USA. It uses 504 MWh on average daily, summing up the energy consumed by around 17 thousand average homes in the U.S. daily. And this is only one data center.
Now, regardless of these numbers being too high, they are considered a success in efficiency. Digital engineers have created smaller and more efficient transistors, improved the circuits, the software, and the power-management schemes. But despite these tremendous improvements, the workload has also increased so that more enterprises need more and larger data centers with an annual energy consumption growth of 20-40%. For instance, the combined electricity demand of Amazon, Microsoft, Google, and Meta doubled between 2017 and 2021. Besides, the smaller-size transistors have been presenting current leakage problems, posing significant challenges, causing the chips to heat up, inevitably increasing the energy consumption.
Lowering the energy footprint with quantum computers
Now, how about looking outside traditional computing? Quantum computing is rapidly emerging as an up-and-coming next generation of high-performance computing to address complex problems inaccessible to classical devices. Many scientific and industrial computations scale exponentially in time on classical machines.
These kinds of problems are called "intractable" because their calculation time increases in unreasonable proportions with their size. Common examples are optimization problems: finding the best option, according to determined criteria, out of a usually tremendous number of combinations. For instance, an optimization problem could be the best way to deploy a communications network, reaching good coverage and, at the same time, lowering the cost. Other typical examples of intractable problems are chemistry simulations at the molecular level, which are particularly crucial in healthcare, such as drug design or toxicity prediction.
The promise of quantum technologies to tackle such intractable problems in a "human scale" amount of time is called quantum advantage.
The question is, will quantum advantage arrive with energy consumption advantage?
Today, quantum computers' electricity usage is orders of magnitude much less than any supercomputer, and this is counting all the different quantum architectures available. Let's take for example superconducting qubits, the most expensive architecture, and these computers only consume about 25 kW. That amounts to 600 kWh daily, a thousand times less than the Frontier supercomputer. Much less is the consumption of neutral atoms quantum devices, such as PASQAL's, which amount up to 7 kW.
Although quantum computers have proven superiority to classical computers in tackling particular scientific problems, they are not yet ready to solve industry-relevant problems. This happens because current quantum computers are noisy, and fault-tolerant quantum computers won't be available for a while. However, because some quantum algorithms are designed to successfully work within the so-called Noisy Intermediate Scale Quantum (NISQ) era, certain architectures, such as PASQAL neutral atoms devices hold the potential to tackle many industry-level use cases before the fault tolerance era arrives. In this scenario, we can foresee that, in the shorter term, there will be, indeed, an energy quantum advantage. Another point to be considered is that quantum technologies are not meant to replace classical CPUs but to collaborate with them. Therefore, the hope is that quantum computers lower the footprint in this hybrid scenario.
Neutral atoms, a greener quantum architecture
Because there is not yet a quantum technology standard, different kinds of architectures are competing in the market. The table below compares energy usage, including the most popular types of quantum computers. From there, you can see that PASQAL’s neutral (or cold)atoms machine is one of the most convenient for energy consumption in total and per qubit.
The energy cost depends on the components each machine needs to work with and, for most of them, on the number of qubits. Usually, the most significant costs are associated with the cryogenic system, which is largely needed in the superconducting qubits type but much less for cold neutral atoms.
The electricity consumption of PASQAL’s current quantum processor, which has demonstrated its capacity to work with hundreds of qubits, shows a total power consumption of around 2,6 kW. Half comes from the lasers; the rest is divided between electronics and environment control. An independent assessment performed on PASQAL technology by Yann Portella and Jolan Tissier from Centrale-Supelec University1 shows that the current neutral atoms PASQAL machine’s power consumption is independent on the number of qubits, contrary to other competing architectures such as superconducting or silicon. These assessments consider that cold atoms devices need a classical computer to optimize the laser operations.
However, adding a 4 Kelvin cryogenic system will increase electricity usage in the next generation machines, with an estimate decrease of about 7 kW of a total consumption of 9,7kW for the 1000 qubits and 9,8 kW for the 10 000 qubits machine, which is still much less energy usage than superconductors. It is important to consider that these projections are rough estimations, there are still uncertainties that will clear up when these devices are actually operating. Besides, we still need to consider the hybrid (classical and quantum) workflow needed by most algorithms.
What will happen in the long term?
The long-term quantum technologies energy usage is unknown, but some estimations have been modeled considering scalability and error correction techniques, which will require large amounts of qubits, enabling a decent time to perform calculations at a low error rate.
The figure below, provided by Alexia Auffèves and Olivier Ezratty, 2022, shows a comparison of energy consumption in terms of the size of a classical intractable problem. There, we can see that even with the introduction of error corrections, we will still have an interesting energy advantage for quantum computers.
There are still more questions that need to be addressed, for example, if this potential energy advantage of quantum computing will hold for all kinds of algorithms and applications. Also, it is important to consider the greenhouse gases footprint corresponding to the fabrication of the computers in both cases, classical supercomputers and quantum computers. Other questions come from the economic and societal issues, such as what will happen when quantum computing becomes widespread and how can it be limited?
Scientists and engineers need supercomputers for calculations and simulations to create more knowledge, manufacture better products, and solve the most pressing problems of our times, such as climate change. But to fight climate change, enterprises offering computing services, whether classical or quantum, should keep researching to lower their energy footprint, while humanity should keep searching for ways to reach our global sustainable development goals.
1We are thankful to Yann Portella and Jolan Tissier from Centrale-Supelec University for their independent analysis of PASQAL technology energy consumption, 2023.
The author is thankful to PASQAL's advisor Etienne de Rocquigny, former Vice-Dean of Research at Centrale Paris and co-founder of Blaise Pascal advisors, for great discussions and feedback.
Are you ready to find new ways to face sustainability challenges for our cities and transport? Such as lowering consumption, increasing energy efficiency, and discovering endless new possibilities. Participating in the Blaise Pascal [Re]generative Quantum Challenge you will have the opportunity to push the boundaries of quantum computing and its applications, influence and [re]shape the world and the future ahead of us.
Would you like to learn more about these techniques on a neutral atom quantum computer? Get familiar with quantum computing, our platform, and algorithms with Quantum Discovery.