Intel Releases Version 1.0 of their Quantum SDK

Intel Quantum Software Development Kit, Credit Intel

Last September we reported on Intel's beta version release of a Quantum SDK for developing and simulating quantum programs. Now, they have released a version 1.0 of this software with some new features. The software is based upon the C++ language and starts with the low-level virtual machine (LLVM) compiler toolchain with added extensions for quantum. The software also supports a quantum runtime environment optimized for executing hybrid quantum-classical algorithms.

We asked Intel why they chose C++ instead of Python and their belief is that for those programmers with a classical computer software background and just starting to learn how to program a quantum computer , C++ is a more familiar choice. In addition, from a performance standpoint, C++ is faster than Python. Although that wouldn't affect the actual quantum operations, for compiling and optimizing programs, this could provide an advantage as the program complexity gets larger. In order to accommodate people who have already have Python-based quantum programs, Intel has provided a new bridge feature in this revision 1.0 that can translate a Python application to C++. So, for example, if a user has a library of programs they developed using Qiskit, they would be able to use this capability to compile and run one of these programs on the Intel simulator and eventually, the Intel hardware.

Besides the Python to C++ translator, the second major feature Intel has added to the software is a simulator that emulates Intel's quantum dot hardware and the Horse Ridge II qubit control chip. (This is shown in the middle column in the diagram above). The Intel Quantum Simulator that was released in the beta version last year is still available and can support 32 qubits on a single node and more than 40 qubits on multiple nodes using OneApi DevCloud, Intel's HPC cloud service that hosts the SDK.

Intel's quantum team has a full roadmap of enhancements and features for both their hardware and software. They have stated they will be making their quantum hardware available later this year including the Horse Ridge II quantum control chip and quantum spin qubit chip. Details of this have not been provided yet, but Intel did indicate that their SDK will be integrated and work seamlessly with their hardware when this happens. (The third column in the diagram above.) Intel is also indicating they will be continuing to develop and add new features into the SDK with future releases. Some of the capabilities may include features like noise modelling, pulse level control of the qubits, application libraries, and others.

Intel's overall quantum strategy is a long term one. They believe that there won't be significant production use of quantum computers until machines with large numbers of qubits and error correction are available. They believe this will take many years to come to fruition, but their choices of technology today are guided by that vision.

For more information about Intel's new release of their quantum SDK, you can access a Newsbytes article on their website here and also a blog post that describes some work being done by beta users of the software here. For those who want to try out the software, you can start by creating an account on their DevCloud system and registering here.

February 28, 2023