This article is more than 1 year old

Assume the superposition: Intel emits SDK to simulate quantum computers

32 virtual qubits are yours to manipulate with classical C or Python code

Intel has released a Quantum software development kit (SDK) that simulates a complete quantum computer using conventional hardware.

Chipzilla's released the kit in recognition of the fact that the few working quantum systems from the likes of IBM or D-Wave, are prohibitively expensive to buy or rent for early quantum experiments.

The silicon slugger therefore hopes the SDK allows developers to start experimenting with quantum algorithms and how they can be integrated into existing platforms, with an emulator.

"It will not only help developers learn how to create quantum algorithms and applications in simulation, but it will also advance the industry by creating a community of developers that will accelerate the development of applications," Anne Matsuura, director of Quantum applications and architecture at Intel Labs, explained in a statement.

Proponents of quantum computing believe the technology will deliver machines so powerful they'll make classical computers look like slide rules, allowing scientists to swiftly make breakthroughs in fields such as encryption, real-time pathfinding, and drug discovery.

While that sounds lovely, existing quantum computers are hard to build, harder to operate, and even the most advanced quantum systems – IBM's Osprey, for example – might have a few hundred qubits, well short of the hundreds of thousands of logical qubits required to achieve the promised revolution.

Another challenge – which Intel's SDK seems to target – is finding practical applications for quantum computers. Any hardware is only desirable if there's software to run on it, and at present humanity's knowledge of what works well on a quantum machine is limited, hype about drug discovery and all that notwithstanding.

Intel claims its Quantum SDK allows developers to get a head start by letting them build software while engineers figure out how to put actual qubits into working and accessible quantum computers.

The SDK is written in C++ using a low-level virtual machine compiler, which the company says makes it easier for developers to simulate quantum environments and integrate them into their existing C, C++, and Python applications.

Intel's Quantum Simulator (IQS) – an open source project that's been kicking around for a few years now – provides a backend for the SDK. IQS simulates generic qubits, which in theory should allow developers to port their work to real quantum systems in the future.

If any of this sounds familiar, it's because Intel isn't the only company trying to emulate quantum systems using conventional platforms. Last year, Fujitsu claimed it had develop the world's fastest quantum simulator, capable of handling 36 qubit quantum circuits.

However, as anyone who's dabbled with hardware emulation will know, simulating logic in software can be incredibly inefficient. This certainly was the case for Fujitsu's simulator, which required a 64-node cluster of its PRIMEHPC FX 700 boxes, each powered by the Arm-based  A64FX processor at the heart of the Fugaku supercomputer.

By comparison, Intel's IQS supports simulations up to 32 qubits on a single node – or can be scaled out to multiple nodes to emulate larger systems. The SDK will eventually work with Intel's own quantum hardware, including the Horse Ridge II control chip and upcoming quantum spin qubit chip.

Intel admits there's still work to be done to fully integrate the SDK with its quantum hardware – assuming that CEO Pat Gelsinger doesn't take the ax to the project as he attempts to turn around Chipzilla's wonky finances. ®

More about

TIP US OFF

Send us news


Other stories you might like