High performance computing (HPC) is commonly applied by organisations to use-cases of scientific, industrial, and societal importance in cases where the need for greater computational and data-driven capabilities are required. The ability to accelerate drug discovery, develop new materials at the molecular level, comprehend the impact of climate change or understand fine grained relationships in vast amounts of data would not be possible without HPC. Since the advent of the first mainframes and supercomputers, the HPC community and industry has always been at the forefront of technology and continues to play a pivotal role in the development, adoption and maturation of novel technologies. New computational capabilities allow for the solution of problems of greater complexity, which in turn lead to the development of further novel technologies and increasingly sophisticated applications. These capabilities are applied by organizations around the world every second of every day on systems located in data centers, supercomputing centers, and the cloud.
High Performance Computing for Simulation and AI
Traditionally HPC architectures were designed for simulation-based approaches such as air flow simulation, seismic imaging and portfolio optimization. The next major development was with artificial intelligence which has made significant advances over the past decade. The availability of sufficiently rich datasets combined with the advancement of machine learning techniques and HPC computational power have made possible the application of AI to a wide range of areas.
Today’s HPC systems are characterized by the convergence of simulation and AI. Systems architectures are highly complex composed of multiple processors and computational accelerators, memory, communication, and storage technologies. This heterogeneity reflects the varied characteristics of applications which have evolved to leverage multiple techniques, such as numerical simulation and machine learning, to reach the greatest levels of performance and information processing. This convergence of techniques and technologies has enabled the solution of problems not previously possible and the formulation of entirely new applications.
The next step is Quantum Computing: HPC-QC
The race to commercial quantum computers is one of the most exciting technological challenges of our generation. The potential to disrupt the status quo in unprecedented ways is fueling nations and organizations across academia, government, and industry to accelerate its realization and application to problems of socio-economic importance. We are at an inflection point where organizations and countries are investing in quantum technologies with the ambition to incubate domestic industries and to support long-term resilience and growth.
While quantum computing will not replace classical computing, the goal of quantum computers is to be able to address questions that are either intractable or impossible by their classical counterparts. The power of quantum computers stems from the most fundamental rules of quantum mechanics in which qubits can be in many states at the same time. Today’s quantum industry is still quite nascent with many quantum processor architecture modalities and an evolving technology stack. Much is still unknown about how large-scale quantum systems will be architected to achieve fault tolerance and commercial viability. Researchers are exploring the immense potential of quantum algorithms and applications which will require rethinking approaches to implementation with the opportunity for the development of entirely new use-cases. Applications and algorithms that rely on probability, factoring and optimization, such as supply chain management in logistics, risk modeling in financial services, molecular engineering in drug design, nano design of materials, fraud detection in cybersecurity and image recognition in artificial intelligence, have been identified as applications that could leverage the exponential power of quantum computers. Problems that could take days, months, or years with classical approaches, could be solved in a fraction of the time with a quantum capability.
As with HPC convergence for simulation and AI, we can expect a similar trajectory with quantum computing where quantum accelerators will become integral parts within larger scale HPC systems. HPC is already playing a key role in the deployment of the first significant quantum capabilities at HPC centers around the world. While quantum computing has special functional characteristics unlike any other technology, the effective integration of quantum algorithms alongside simulation and machine learning will be essential to its broad adoption.
As industry and society seeks to understand the world around us, HPC will continue to play a vital role in addressing challenges of socio-economic importance. Quantum computing is a generational capability that will extend the capabilities of HPC. What is certain is that quantum computing is a radically different approach that requires specialist skills and fundamental rethinking of applications. As such, equally important to the advancement of technology is the development of the workforce to leverage it. Continuous access to leading edge HPC-QC facilities is an essential ingredient to the development and incubation of technologies and skills.