IDC Perspective on Integration of Quantum Computing and HPC – HPCwire

The insatiable need to compress time to insights from massive and complex datasets is fueling the demand for quantum computing integration into high performance computing (HPC) environments. Such an integration would allow enterprises to accelerate and optimize current HPC applications and processes by simulating and emulating them on todays noisy intermediate scale quantum (NISQ) computers.

Currently, enterprises are reliant on the advantages that can be achieved using only classical accelerator technology such as GPUs and FPGAs. However, HPC systems are limited in their ability to process and analyze large amounts of data needed to execute multiple workflows, even with the added compute power of classical accelerators. Using quantum computing technologies, not only will enterprises be able to accelerate current HPC processes, but they will also be empowered to solve intractable industry problems beyond the scope of the most advanced classical compute systems.

Today, quantum computing systems are still in early development and far from commercial maturity. Quantum computing hardware vendors are challenged in their ability to stabilize and scale the large number of qubits needed to solve complex problems and allow for error correction due to decoherence. As a result, NISQ machines cannot provide a means for enterprises to realize a quantum advantage, defined by IDC as being able to solve a problem that has actual value to a business, humanity, or otherwise.

Despite these challenges, enterprises are investing in quantum initiatives to identify uses cases and develop algorithms so that they are quantum ready when a fault-tolerant universal machine is realized. As a result, government entities, such as China, Germany and the US; IT industry leaders such as IBM, Google, Microsoft, and Amazon Web Services (AWS); and private investors are escalating funding for quantum computing to push this technology to new levels of maturity.

IDC expects investments in the quantum computing market will reach nearly $16.4 billion by the end of 2027. IDC believes that these investments will lead to waves of technology innovation and breakthroughs that will allow organizations to apply quantum computing to a diverse and expanding group of use cases that involve the analysis of huge amounts of diverse datasets, exponentially large numbers of variables, and an inexhaustible number of possible outcomes.

The ability to address large-scale use cases using quantum computing is possible due to the qubits unique superpositioning and entanglement properties. Quantum and classical computers store and compute data based on a series of 0s and 1s. In classical computing, this is done using a bit. Bits are only capable of holding the values of 0 or 1. Bits cannot hold the value of 0 and 1 simultaneously. Qubits do have this capability. This property is referred to as superposition. Through qubit entanglement, a pair of qubits is connected or linked. Change in the state of one qubit results in a simultaneous, predictable change in the other qubit. Combined, the quantum properties of superpositioning and entanglement provide qubits the ability to process more data faster, cheaper, and better (more accurately or precisely) than a classical computer. As a result, enterprises can use quantum computing systems to explore new and unique use cases which can accelerate current business processes and workloads.

The list of use cases is growing at a rapid pace. Included in this list are performance intensive compute (PIC) specific use cases that address newly defined problems, refine solutions generated and iterated in the PIC environment, simulate quantum algorithms, and more. Energized by this innovative technology, many enterprises dont want to delay the commencement of their quantum journey. Approximately 8 out of 10 enterprises that are currently investing, or planning to invest, in quantum computing expect to integrate quantum computing technologies as a hybrid model to enhance their current performance intensive computing (PIC) capabilities. Because of this trend, IDC anticipates that several performance-intensive computing workloads will initially be turbocharged by quantum computing-based accelerators. Yet, in the long-term many of these workloads will eventually cross the computing paradigm and become quantum only.

Quantum and classical hardware vendors are working to develop quantum and quantum-inspired computing systems dedicated to solving HPC problems. For example, using a co-design approach, quantum start-up IQM is mapping quantum applications and algorithms directly to the quantum processor to develop an application-specific superconducting computer. The result is a quantum system optimized to run particular applications such as HPC workloads. In collaboration with Atos, quantum hardware start-up, Pascal is working to incorporate its neutral-atom quantum processors into HPC environments. NVIDIAs cuQuantum Appliance and cuQuantum software development kit provide enterprises the quantum simulation hardware and developer tools needed to integrate and run quantum simulations in HPC environments.

At a more global level, the European High Performance Computing Joint Undertaking (EuroHPC JU) announced its funding for the High-Performance Computer and Quantum Simulator (HPCQS) hybrid project. According the EuroHPC JU, the goal of the project is to prepare Europe for the post-exascale era by integrating two 100+ qubit quantum simulators into two supercomputers and developing the quantum computing platform, both of which will be accessible via the cloud.

Due to the demand for hybrid quantum-HPC systems, other classical and quantum hardware and software vendors have announced that they too are working to develop a hybrid quantum-HPC solutions. For example, compute infrastructure vendor, HPE, is extending its R&D focus into quantum computing by specializing in the co-development of quantum accelerators. Because quantum software vendor, Zapata, foresees quantum computing, HPC, and machine learning converging, the company is creating the Orquestra Universal Scheduler to manage task executions on HPC clusters and current HPC resources.

Yet, recent results from an IDC survey indicate that approximately 15% of enterprises are still deterred from quantum computing adoption. For quantum computing to take off, a quantum computing workforce made up of quantum scientists, physicists, engineers, developers, and operators needs to evolve. However, this should not deter enterprises from beginning their quantum computing journeys. Instead, hesitant adopters should take advantage of the development and consulting services offered by quantum hardware and software vendors, as well as IT consultants that specialize in quantum computing technologies. Because the choice is clear, become quantum ready or be left behind. IDC projects that worldwide customer spend for quantum computing will grow to $8.6 billion in 2027.

Authors

Heather West, Ph.D., Senior Research Analyst, Infrastructure Systems, Platforms and Technologies Group, IDC

Ashish Nadkami, Group Vice President, Infrastructure Systems, Platforms and Technologies Group, IDC

Sample of IDC Reports

Worldwide Quantum Computing Forecast, 2021-2025: Imminent Disruption for the Next Decade

IDCs Worldwide Quantum Computing Taxonomy, 2022

Emerging Trends in End-User Adoption of Quantum Computing-as-a-Service Solutions

2021 Worldwide Quantum Technologies Use Case Report

Here is the original post:

IDC Perspective on Integration of Quantum Computing and HPC - HPCwire