Hello and welcome to my blog!
In this
article I will try cover the following.
Please note I use the term QC(s) as an abbreviation for “quantum computer(s)”
- [Introduction/context] How traditional computers work and how are quantum computers different)
- Why do we need QCs? And why traditional computers can’t do the job instead of QCs?
- How fast could quantum computers be?
- Who’s currently developing QCs?
- What is the current usage of the existing QCs?
- What is the intended usage in the future?
- What are the limitations/issues? Why we don’t have large scale production of QCs yet?
- When will we have a commercial (& consumer grade) QCs for sale on large scale?
- Will They replace current computers?
Introduction
In a world full of data and information, the amount of calculations on them keeps increasing and it’s getting even more and more complex. A way to store data and perform actions on this stored data is the classical computers that we all know.
As I explained, Computers are there to answer
our demand for data and calculations. In simple terms, we store data as bits,
electronic signals that can be either High or Low, only one value at a time. We
represent these two statuses with the famous binary model consisting of
sequences of 1’s and 0’s.
We perform operations on these data thanks to a CPU, the central processing
unit (aka. processor) that allows us to perform operations. The faster your CPU
is the more calculations per second it can perform. Fast here refers the CPU
frequency and the number of cores. This translates physically to the quantity
of transistors the CPU has.
In quantum computing, we store data as quantum-qubit (qubits). Those qubits can possibly be on (1) or off (0) or they can even be both 1 and 0 at the same time. Data are processed through a quantum processing unit (QPU). The more qubits the QPU has the faster the computer will perform operations.
Why do we need QCs? And why traditional computers can’t do the job instead of QCs?
So Why bother developing those QCs? Why can’t we
simply keep improving our computers and create 1024 cores processors with 10GHz
each core? Well here’s the thing. Remember when I said: better performance = more transistors? Here
lies the issue. Unfortunately, the number of transistors we can keep adding is
limited by some physical constraints. Like the amount of heat for example. Higher
frequency CPUs will generate more heat, and this will be an issue to keep them
cool. Also, there’s a law called: Moore’s law. This law explains the
limitations of the transistors we can keep adding to our CPUs. According to
this law, we almost reached the limit. This means that we can hardly make
computers any faster.
So that’s it? Is it the end? We still need faster computers to accommodate for
the increasing demand of data processing. Luckily, we came with a couple of
workarounds like cloud computing and parallelism. And another possible
solutions to this problem is quantum computing.
Quantum computing still can’t break Moore’s law. But since it is a law about transistors in a dense integrated circuit it simply does not even apply to Quantum computers (QCs). QCs are not affected by it. However, another law exists for QCs, it’s called rose’s law but we’re still far from reaching its limitations.
How fast could quantum computers be?
In theory, they’re million times faster than our traditional computers. The performance of QCs depends on the number qubits. A 30 qubits machine might offer the same performance expected from current computers. Here’s a quote from quantumly
In 2015, Google and NASA reported that their new 1097-qubit D-Wave quantum computer had solved an optimization problem in a few seconds. That’s 100 million times faster than a regular computer chip. They claimed that a problem their D-Wave 2X machine processed inside one second would take a classical computer 10,000 years to solve.
Who’s currently developing QCs?
The number of companies, labs and universities working on quantum computers keeps increasing every day. Here’s a non-exhaustive list of the ones I know in various Countries:
- North America
- USA:
- Nasa/ QuAIL
- IBM
- Microsoft
- Intel
- Mit
- Caltech
- UC berkley
- Canada
- D-wave
- IQC
- USA:
- Europe
- European non country-specific:
- OpenSuperQ (Germany, Spain, Sweden, Switzerland)
- Quantum Technologies Flagship
- United Kingdom
- Cambridge Quantum Computing
- UKNQTP
- UCLQ
- Ireland
- Accenture
- France
- Atos
- Pcqc (/cnrs)
- Switzerland
- NCCR QSIT
- Netherlands
- Tu delft/ qutech
- Germany
- TUM university
- European non country-specific:
- Asia
- China
- Baidu
- Japan:
- Tokyo Tech laboratory
- University of Tokyo
- RIKEN,
- The National Institute of Advanced Industrial Science and Technology
- Honk-Kong:
- HKUST
- China
There’s still much more of course and the list will keep increasing but these are the ones I found at the time of writing.
What is the current usage of the existing QCs?
Quantum computers are great for solving optimization problems and they can do even much more. Due to the limited access, we still didn’t fully utilize them, they’re only used for a limited number of tasks like the following:
- machine learning, pattern recognition, mission planning and scheduling, distributed navigation and coordination, and system diagnostics and anomaly detection (according to NASA).
- Volkswagen recently used QC from D-wave to help optimize traffic routing. (source)
- Most other uses at the time being are not real-world use cases, they’re mostly research oriented making use of quantum computing to run scientific simulations.
Here’s an article talking about some uses of the existing quantum computers.
What is the intended usage in the future?
In the future, if large-scale fully usable quantum computers are built, they will be able to solve certain problems exponentially faster than our current classical computers (for example Shor’s algorithm). Quantum computers may be used to break all the cryptographic encryptions we have today. Similarly, they can be used to create “unbreakable” encryption systems that are much more secure and almost impossible to break even with another quantum computer. That said, quantum computer can help introduce a huge improvement in various domains that require calculations such as:
- Cybersecurity
- Much more accurate weather forecast
- Drugs development.
- Banking and financial modelling
- Artificial Intelligence and machine learning
- Physical simulations of scientific experiments.
However, it’s worth noting quantum computers may not be beneficial (or at least have not been proven of use) to all domains. For example, domains like sound/video editing, graphic designing and video games won’t see any improvement. Video encoding and compression might be however improved since it’s an algorithm-based process.
What are the limitations/issues? Why we don’t have large scale production of QCs yet?
In the meantime, we have a couple of issues that we still need to solve in order to attain a reliable QC. To do this, we need to find an optimal “quantum error correction”, it’s the mechanism used to prevent Quantum decoherence and Quantum noise. It’s worth noting that noise may be caused by temperature fluctuations, mechanical vibrations or stray electromagnetic fields. In All cases the noise weakens the correlations between qubits and may result in errors in the data. To increase reliability, QCs are cooled with huge cooling systems making them room-size computers. The size of the D-wave 2000q is around 10′ x 7′ x 10′.
When will we have a commercial (& consumer grade) QCs for sale on large scale?
I luckily had the chance to talk to some researchers from CNRS and I asked them this question. According to them, we expect to reach production ready quantum computers on much larger scale by 2030. I should mention that since 2017 we already have access to IBM’s network of quantum computers via their cloud services.
Will They replace current computers?
As I said earlier, domains like sound/video editing, graphic designing and video games won’t benefit from quantum computing (except for the computational part but not the visual part). QCs might be even worse than classical ones at some tasks. The current quantum computers we have are all hybrid so they can communicate with our normal computers. This means that the quantum computer containing the QPU is connected to another conventional computer that is programmed specially with all necessary functionalities to control the quantum computer connected to it. Access to the quantum computer is done by accessing the conventional one connected to it (locally or remotely) and the computer will then perform the calculations on the QC that is attached to it.
In short, QCs were never meant to replace today’s computers and they won’t replace them.
Disclaimer: I try to put references to my sources as much I can, but a lot of what I write is based on my understanding of what I read in various articles. My articles should not be used as a source of information for any research projects or thesis as they may possibly contain some mistakes. If you ever spot any mistake in any of my articles, please do not hesitate to let me know!