About Marcus Köhler
About Marcus Köhler
Report 3/25 15 min read
​​​
​
​
​
​
​
​​​
The data center market is growing rapidly worldwide, driven by the increasing demand for computing power for cloud services, artificial intelligence, and big data. At the same time, the high energy consumption of data centers is a key challenge. Deloitte Insights (2022) (*1) highlights that sustainable alternatives are urgently needed to meet environmental and economic challenges. The increasing demand for computing power will further exacerbate this trend in the coming years.
With advances in Quantum computing, a disruptive technology could emerge that will permanently change the data center investment market. Recent studies by McKinsey (*2), Gartner (*3) and IBM Research (*4) show that Quantum computing will have a significant impact on IT infrastructure, particularly in the areas of energy efficiency, investment shift and hybrid computing architectures.
​​
Power consumption and computing power of classical computers vs. Quantum computers (*5)
Quantum computers can be exponentially faster than classical supercomputers for certain problems (e.g. optimization, cryptography). However, classical computers would require many times more energy to perform the same highly complex calculations.
​
Classic supercomputers
Computing power: Fugaku, currently the most powerful supercomputer, achieves about 442 petaflops (442 quadrillion floating point operations per second).
Power consumption: Fugaku requires about 30 megawatts (MW) (*6).
​
​Quantum computers (IBM, Google, D-Wave)
Computing power: No direct FLOPS figure possible, but Google's Sycamore is said to have performed a calculation in 200 seconds that would take a classical supercomputer 10,000 years.
​
Power consumption
The exact power requirements of Google's Quantum computers, such as the Sycamore processor or the newer Willow chip,
are not publicly disclosed by Google. However, it is generally known that quantum computers have significant power
requirements due to their complex cooling and operating systems. A 2024 study reported that the Sycamore processor used
about 4.3 kilowatt hours (kWh) in 600 seconds (10 minutes) for a given calculation, which is an average power consumption of
about 25.8 kilowatts (kW) (*7). IBM's Quantum computer (Eagle with 127 qubits) uses about 10-20 kilowatts (kW) (*8).
Efficiency Comparison
A direct comparison of energy efficiency between classical computers and Quantum computers is difficult because the computational methods are fundamentally different. While classical computers are based on bits with the states 0 or 1, Quantum computers work with qubits and complex probability distributions. Nonetheless, it is clear that Quantum computers offer unparalleled computational power for certain applications with comparatively very low energy consumption. The energy requirement is almost negligible compared to the achievable computing power. However, the high demands on cooling and stability are still major technical challenges that need to be solved.
​
Energy needed to cool Quantum computers
Quantum computers must be operated at extremely low temperatures because the qubits can only operate reliably in an almost interference-free state. Most existing quantum computers are based on superconducting qubits, which require temperatures close to absolute zero (about 15 millikelvin or -273,135°C). This requires sophisticated cooling systems that require significant amounts of energy:
Dilution refrigeration systems: These cooling systems use a mixture of helium-3 and helium-4 to achieve the extremely low
temperatures. The process of diluting the helium isotopes continuously removes heat from the system.
Multi-stage cooling process: The cooling process takes place in several stages, starting with commercial compression cooling systems at 50 Kelvin (-223,0°C) to helium cryocoolers at 4 Kelvin (-269,0°C) to final dilution cooling.
Energy requirements for cooling: IBM and Google estimate that the cooling infrastructure alone for a large quantum computer currently consumes about 10-20 kW. While this figure is relatively low compared to conventional supercomputers, it represents an additional energy challenge.
Alternative cooling technologies: Research facilities are investigating alternative cooling methods, including laser cooling,
superconducting materials with more stable qubits, and new cryogenic cooling systems that could reduce the energy required
for cooling.
​
Why Quantum computers will not completely replace conventional computers
Although Quantum computers are revolutionary for certain tasks, there are several reasons why they will not replace all conventional computers in the long run:
Cost: Quantum computers are extremely expensive to build and operate. The infrastructure requires specialized hardware,
cryogenic cooling and high-precision control systems, which limits their widespread use.
Complexity and need for specialists: Quantum computers require specialists to program, maintain and operate them. The
algorithms are fundamentally different from those of conventional computers, making the transition difficult across the board.
Specific applications: Quantum computers are particularly efficient at optimization problems, simulations and cryptographic
computations. For everyday applications such as office software, databases or simple calculations, conventional computers
will remain more efficient.
Existing investments in traditional computing technologies: Companies have already invested billions in traditional IT
infrastructure. Complete replacement by Quantum computers would not be economically viable.
Scalability: Current Quantum computers have a limited number of qubits and are highly error-prone. Until a universal Quantum
computer with sufficient stability exists, hybrid systems will dominate.
Coexistence of Quantum and conventional computers
Despite the enormous progress in Quantum computing, it is unlikely that it will completely replace conventional computers in the near future. Instead, a hybrid structure will emerge in which both technologies coexist. McKinsey predicts that by 2035, about 20% of the most demanding calculations will be performed by Quantum computers, while conventional computers will continue to be used for general applications.
Conclusion
The success of Quantum computing will have a profound impact on the data center investment market. While investments will continue to be made in traditional data centers in the short term, hybrid models and quantum computing infrastructures will become increasingly important in the long term. The challenge for investors will be to set the course early on for a future in which both technologies are optimally leveraged. Ultimately, Quantum computing also offers a promising opportunity to balance the growing demand for computing power with sustainable solutions.
​​Sources
-
own Koehler Advisory research
-
(*1) Deloitte, https://www2.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions/quantum-computing.html
-
(*2) McKinsey, https://www.mckinsey.com/business-functions/mckinsey-digital/our- insights/the-next-tech-revolution-quantum-computing
-
(*3) Gartner, https://www.gartner.com/en/insights/quantum-computing
-
(*5) Springer Nature, https://www.nature.com/npjqi/
-
(*6) Top500, https://www.top500.org/lists/top500/
-
(*7) Cornell University, https://arxiv.org/abs/2407.00769?utm_source=chatgpt.com
-
(*8) IBM Research, https://research.ibm.com/quantum-computing

Impact of Quantum computing on DC market