Computational thinking and working makes its entry into the business world. The use of new concepts such as computer simulation techniques, machine learning, artificial intelligence (AI) or internet of things (IoT) will enable more and more products to be replaced by services, that individualized products will be offered and this in a hyper-competitive world with a shortage of certain goods and labour.
These new concepts include a wide range of digital possibilities from networks across storage and computing power to smart communication options and brand new software solutions. These are concepts that are also present in the HPC world, are monitored and deployed in an innovative and thorough way. For different sectors in this digitized economy, the use of large-scale computing capacity, large-scale storage capacity, powerful connectivity and innovative software, in short high performance computing, "the" tool to improve products and services in their domain.
Although manufacturers have been using supercomputing for many years, the industry’s needs are still growing rapidly. Especially in the area of computational fluid dynamics (CFD) workloads, HPC is highly needed. CFD acceleration is applicable in many different markets, such as automotive, construction, consumer goods, energy, aerospace etc. By modelling, simulation and data analysis, supercomputer applications can address key challenges in manufacturing and material development, which results in an improvement of energy efficiency, increase of productivity, reduction of the cycle time, acceleration of innovation, and more.
In the energy industry, HPC aims to apply supercomputing techniques to for example energy industry simulations. These simulations are being customized and go beyond the state-of-the-art for several energy sources, like wind energy production and design. As another example, supercomputing already helps the oil and gas operations from reservoir exploration and development to the operations of the energy assets. This includes upstream production and refining. As a conclusion, numerical simulations has still a huge potential to become one of the essential tools for the development of new innovative solutions for energy production and conversion and modernization of existing equipment.
Cyber Security & AI
The amount of internet traffic data generated in short periods of time is massive. As a result, it is impossible for one or even 100 laptops to process this data into something that human analysts can digest. Therefore, analysts rely on sampling in their search for potential cybersecurity threats: they select small segments of data which the examine in depth, hoping to find suspicious behavior. Supercomputing is still an untapped resource in this domain and has the potential to disrupt its properties.
In the domain of A.I., there are already efforts to explore the potential of training complex deep neural networks (DNNs) on supercomputers, which could be used in numerous applications. Cybersecurity is one possible application, but there are many more.
For years, numerical modelling and simulation has been used in aerospace to create high-end products. This way, supercomputing has made significant contributions in aerospace engineering in recent decades, such as advances in computational fluid dynamics (CFD), and so altering the way aircrafts are designed nowadays. Furthermore, the relentless growth in supercomputing power holds promise of huge leaps in engine performance and other aerospace technology.
Because of the great amount of data that is being analysed and the sophisticated analytical workflows that are being applied to large cohorts of data, supercomputing has already become a mainstream requirement in biomedicine nowadays. Currently, there are mainly six major domains in the biomedical industry that are using supercomputing: genomics, imaging, molecular dynamics & computational chemistry (simulations), structural biology, proteomics, and math and statistics. However, there are still many innovations that can be made and many new applications to be supported by HPC design. Examples are real-time feedback computing during complex surgical procedures using detailed imaging techniques, as well as hospital laboratory diagnostics that return instant results and personalized treatment regimens (precision medicine).
Also in financial organizations, supercomputing is already in use in very different ways. Examples are:
Portfolio optimisation: running models and optimising thousands of individual portfolios overnight, on the basis of the previous day’s trading results.
Valuation of financial derivatives: for example, a re-insurance firm might need to value and compute hedge strategies for hundreds of thousands of policy holders in its portfolio.
Detection of credit card fraud: supercomputers enable a bank to easily run more fraud detection algorithms against tens of millions of credit card accounts.
Hedge fund trading: supercomputers make it possible to react faster to market conditions, enabling analysts to evaluate more sophisticated algorithms that take into account larger data sets.