Cloud-based Quantum Computing
All I know about Quantum Computing is that it is faster. The theory of it gets people smarter than me talking excitedly for long periods of time and I never seem to be any closer to getting a handle on what it means. Now though it seems that the age of Quantum Computers might be drawing near. Google and NASA have claimed they achieved Quantum Supremacy and the big cloud providers are offering access to Quantum Computing, so it is time to get some understanding of what it is and how I can make use of it.
First the basics; What is Quantum Computing?
After looking at a few pages on Wikipedia I came up with the following definition. You can quote me if you like, or if you are smarter than me and found a flaw in my definition then let me know in the comments section and I will quote you.
“A quantum is the minimum amount of any physical entity involved in an interaction. It could be a photon of light or an electron of atomic matter. A quantum can be measured (quantized). Quantum Computing is when you use the quantized state of quanta (plural of quantum) to perform a calculation”
Classical computers use mechanical gates which are either open or shut to perform calculations. Each gate represents one “binary digit” or “bit”.
Quantum Computers use the quantized state of quantum to do the same thing. Each quantum evaluates to either 1 or 0 depending on it’s state. Each quantum evaluation represents one “qubit”.
However, the state of the qubits is not as certain as the state of a gate in a microprocessor and that is the biggest challenge of Quantum Computing. The qubits are considered to have a probability to be 1 and a probability to be 0. The state with the higher probability will be the state it is determined to be in. That’s it, that’s quantum computing.
This all sounds like you will be asking a piece of cheese what the time is. How is this actually going to work?
Despite the hype Quantum Computers are not yet a practical reality. As the state of the quanta is not as stable and reliable as gates on a microprocessor Quantum Computers are prone to err. Quantum computing devices have been created but the errors are still a problem that make them of limited functional value. Fault-tolerant quantum computers are believed by most to be quite a way off yet.
Quantum computing today, including cloud-based Quantum Computing is mostly about running Quantum Computing simulators on classical computers to develop algorithms and programs that will eventually be run on quantum computers when they become viable. The simulators are also used to validate the computations of current quantum computers to assess their fidelity (correctness) to gauge how close we are to having machines that work in practice.
Google AI Quantum team
Google has established a research team focused on quantum computing. The team is a part of the Google AI group and is called Google AI Quantum.
The team made quite a splash in October 2019 when in collaboration with NASA they published a paper in Nature claiming to have achieved “Quantum Supremacy”, meaning they were able to use a quantum computer to solve a problem that conventional computers could not have slved.
The paper abstract states that “A fundamental challenge is to build a high-fidelity processor ..” followed by “Measurements from repeated experiments sample the resulting probability distribution, which we verify using classical simulations”. In other words the experiment validates the results from the quantum computer using simulations in conventional computers.
The abstract goes on to make the big flag-staking claim of “Quantum Supremacy” success:
Our Sycamore processor takes about 200 seconds to sample one instance of a quantum circuit a million times — our benchmarks currently indicate that the equivalent task for a state-of-the-art classical supercomputer would take approximately 10,000 years. This dramatic increase in speed compared to all known classical algorithms is an experimental realization of quantum supremacy for this specific computational task, heralding a much-anticipated computing paradigm. — Arute, F., Arya, K., Babbush, R. et al. Quantum supremacy using a programmable superconducting processor. Nature 574, 505–510 (2019). https://doi.org/10.1038/s41586-019-1666-5
This achievement albeit limited in scope is likely to become the milestone when Quantum Computing became generally considered an inevitable reality.
The team has developed a number of Python libraries for developers to get started with Quantum computing. Published on GitHub under the account “QuantumLib” these open source libraries will get you started with Quantum Computing right on your desktop.
- Cirq; a Python library for writing, manipulating, and optimizing quantum circuits and running them against quantum computers and simulators.
- OpenFermion; an open source library for compiling and analyzing quantum algorithms to simulate fermionic systems, including quantum chemistry.
Google has also published a “Quantum Computing Playground” for learning and playing with Quantum simulators running Quantum algorithms.
The Google AI Quantum team has done a lot to progress the evolution of Quantum Computing. In their mission statement they say “We want to offer researchers and developers access to open source frameworks and computing power that can operate beyond classical capabilities”. Doubtless we will be seeing more from them in the future.
IBM Q Network
IBM has a Quantum initiative. It is called IBM Q Network and it’s website is full of management speak.
IBM Q Network is a community of Fortune 500 companies, academic institutions, startups and national research labs working with IBM to advance quantum computing.
.. and ..
We are committed to accelerating and scaling quantum computing by partnering with industries and fostering a growing ecosystem.
Behind the dark suit of enterprise jargon however there is an ambitious Quantum initiative under way at IBM. They have created Quantum Computing Hardware, established a web based quantum computing simulator and developed a number of Python libraries under the GitHub account Quiskit.
The Qiskit initiative even has it’s own website at https://qiskit.org/ and it looks like a lively community.
Quite frankly looking at the artifacts and progress made by the Quantum people at IBM I am surprised that they allowed Google to steal their thunder by declaring Quantum Supremacy. In fact IBM scientists have debated the claim made by Google but marketing wise IBM have lost that round.
For those that have little patience for Qiskit IBM have provided a nice little Web Client called “IBM Quantum Experience”. This will allow you to create algorithms and circuits using a visual interface and cloud based Quantum Computer Simulators.
Through a service subscription on their website you can then run your code on an actual Quantum computer.
Beyond this IBM also has an initiative called “IBM Q System One”. This is presented as the first Commercial Quantum Computer offering. To explain how it works they have provided this press release. This offering appears to be aimed at large research facilities and intended to build up the Q Network.
AWS Braket
Amazon Web Services offer a Quantum Computing service called Braket. It is a managed service which helps you get started with quantum computing by providing a development environment to explore and design quantum algorithms, test them on simulated quantum computers, and run them on quantum hardware provided by partnering companies.
The service is not operational as yet but you can sign up for a preview. The hardware providers are IONQ, D-WAVE and Rigetti.
Q# and QDK - Microsoft’s Quantum Development-Kit and language
Microsoft has taken to the challenge of preparing for the age of Quantum Computing by releasing a development-kit including a new computer language they call Q#.
Microsoft supplement the kit with a cloud based Quantum service. It is not clear from their site whether they only offer simulators or if you can run your code on actual quantum computers on Azure but their partners certainly develop such hardware
Quantum Computing Simulators
There are a number of libraries that will allow you to simulate Quantum Computing on your own computer. In addition to Googles QuantumLib contributions on GitHub have a look at Quantiki’s list of simulators.
The number of simulators available is surprising, there are quite a lot of developers who have been working on solutions in this space.
Conclusion
The age of Quantum Computing is not quite here yet but it’s dawn is looking very near. Research teams are coming closer to creating reliable devices to perform quantum computations and the big vendors are readying their service and product offerings for when that will happen.
The landscape is confusing at present as the vendors race to prepare for a new age of computing that is not yet well understood. Vendor websites are not clear in their statements about product offering what actually is on offer.
However, IBM is probably offering the most cohesive products and solutions. If you are looking to get started with Quantum Computing in the cloud I would start there.
References
Definition of a Quantum
https://en.wikipedia.org/wiki/Quantum
Explanation of Quantum Mechanics
https://en.wikipedia.org/wiki/Introduction_to_quantum_mechanics
Quantum Computing
https://en.wikipedia.org/wiki/Quantum_computing
Wiki article on Cloud-based Quantum Computing
https://en.wikipedia.org/wiki/Cloud-based_quantum_computing
Quantum Supremacy
https://www.nasa.gov/feature/ames/quantum-supremacy, https://doi.org/10.1038/s41586-019-1666-5
Google AI Blog on the Quantum Supremacy achievement;
https://ai.googleblog.com/2019/10/quantum-supremacy-using-programmable.html
