Data-intensive research and the use of advanced digital resources are key to addressing research grand challenges of our age, for the benefit of science, society, and industry. This article focuses on industrial opportunities, including potentially neglected applied research topics, and how access to high-performance computing (HPC) can provide companies with a competitive advantage.
Opportunities for companies to leverage HPC abound now more than ever thanks to the explosion of data being collected. When hearing companies discuss their interests and needs, I’ve heard many if not most say something along the lines of “collecting the data is easy but we don’t know what to do with it.” Whether their challenges are defined as big data not quite requiring HPC or massive data requiring HPC, the opportunities are similar:
- AI: leveraging data to develop definable artificial intelligence solutions to known challenges. This can be simple, surface-level AI through sophisticated analysis and intelligence.
- ML: leveraging data, particularly statistical data, to create machine learning, where a knowledge base grows when fed and analysed.
- DL: leveraging data for broader machine learning – deep learning – that goes beyond the statistical, structured analytics, venturing sometimes into less structured, less predictable learning.
- Other analytics: prescriptive (what to do), predictive (what might happen), and descriptive (what’s happening) analytics are utilised either partially or entirely by companies everywhere.
These major data-driven solutions have existed for years. Because of today’s data explosion, all continue to be opportunities as we all grow in our sophistication of being able to leverage collected data for the enhancement and growth of business operations and offerings.
Lesser known or, perhaps, neglected opportunities to leverage HPC exist as well and are not as obvious. While some companies seek solutions in these areas, most do not, possibly because of lack of knowledge, information, and promotion. These opportunities include:
- Geospatial: Sophisticated, all-encompassing answers to the wide range of questions pertaining to “where” can provide companies intelligence that greatly enhances operations. Think of overlaying geospatial data with severe weather risk alone. Now think of how that can pertain to anything from insurance companies and their claims (tornados, floods, et al.) to manning oil rigs (hurricanes, winds and - like the photo that accompanies this blog) to growing crops (heat, moisture) to consumer preferences. Knowing what’s happening and where is huge data wrought with intelligence and opportunity. It’s also global, as multinational companies can leverage data from around the world to improve operations, safety, outreach, and more.
- Cybersecurity: Obviously, this pertains to protecting your data. When it comes to HPC, you can’t protect systems with antivirus software. Due to the sheer size and scope of data, threat detection solutions can provide high levels of security. In industry, that can mean protecting financial data, personal health information, and government intelligence, among other areas. Most companies do not understand their security options, particularly in HPC, though it should be a top priority, which is why this is such a significant opportunity.
- Other cyberinfrastructure: A term coined by the U.S. National Science Foundation (NSF), cyberinfrastructure is about advanced hardware, networking, and communications to support, in our case, applications that utilise HPC. This is where a real boom and continuous change are happening. It’s an arms race of sorts to build bigger, more powerful and capable machines. It’s faster interconnectivity. It’s a chase for software solutions to keep up with continually advancing compute power and resource. The opportunity for companies involves keeping up – to match developed or provided applications with the right infrastructure to deliver solutions. It’s not always about having the biggest, fastest, or most powerful compute resource. Rather, it’s knowing customers’ needs and delivering that balance of sophisticated products and services all at the right price. Continually having the right cyberinfrastructure is a moving target that requires constant attention and analysis.
- ROI: Businesses rarely take the time to calculate the return on their HPC investment. With HPC being about gained operational efficiencies and intelligence, ROI is a neglected step and missed opportunity all too often.
Impact on Businesses and Summary
Companies that utilise HPC to solve their grand challenges see results in more efficient processes, greater and more sophisticated business intelligence, and cost savings. With attention toward the previously detailed opportunities comes the potential for improved offerings at lower cost – something every company strives to achieve.
Like with most technology-based implementations, initial costs to achieve these advanced levels of operations, service, and support can seem significant. Predictably, when done right, cost savings and ROI can be significant as well, and in the case of HPC – which it is famous for – that ROI and savings can be at massive scale.