Quantum computers also are prone to errors that require resource-hungry correction procedures.
As a result, businesses preparing to adopt quantum computing develop applications using simulation on reliable, classical high-performance computers (HPC) so they will be ready to run on quantum computers. However, simulations are limited in the size of quantum circuits they can simulate.
In an attempt to alleviate this problem, Nvidia has just launched a suite of software development tools called cuQuantum to enable classical HPCs to simulate larger-scale quantum circuits than is currently possible.
This means businesses can prepare software now that will be ready to run on quantum computers later.
Canada-based Xanadu Quantum Technologies has integrated cuQuantum into PennyLane, its open-source framework for quantum machine learning and quantum chemistry.
Oak Ridge National Lab is using it in a framework for tensor
Quantum computers work best for specific types of calculations, so many future applications are likely to use a hybrid of quantum and classical computers to solve problems. Nvidia is also developing a programming language that lets quantum and classical computers talk to each other.
Known as nvq++, it works using Quantum Intermediate Representation (QIR), a type of low-level machine language that quantum and classical computers can use to talk to each other, making the most of hybrid applications.
Researchers have established the QIR Alliance to focus on the interface between quantum and classical computing to get the best from both of them. Led by the Linux Foundation, members include Oak Ridge National Laboratory, Quantinuum and Quantum Circuits. Researchers at the Oak Ridge National Laboratory are expected to be among the first to use this new software.
Enter Quantum Newsletter
To get the latest quantum computing news, advice and insight, sign up to our newsletter