How do you advance your business’s analytics? The future lies in the unification of AI and HPC practices. A recent study shows how this technology is improving firms’ analytic acumen.
How do you advance your business’s analytics? The future lies in the unification of AI and HPC practices. A recent study shows how this technology is improving firms’ analytic acumen.
Advanced analytics delivers insights beyond charts and numbers, including real-time business scenario modeling and simulation, comprehensive variable identification for risk assessment, and improved accuracy in fraud detection. But are your organization’s analytics advanced? How do you advance more?
The future lies in the unification of AI and HPC practices. This convergence is expected to drive more business agility, innovation, and competitive differentiation.
In a recent study, we commissioned Forrester Consulting to evaluate the current state of HPC and AI infrastructure to see exactly how this technology convergence improved firms’ analytic acumen.
The study included a survey of 464 global, cross-industry decision-makers and AI or HPC practitioners, and three expert interviews. The research found synergies between the teams that contribute value to both data scientists pursing AI and HPC computing experts, especially through utilization of shared infrastructure:
#1. The focus on AI is yielding less-than-hoped-for results.
Many businesses are investing in critical infrastructure to support their AI initiatives. The study showed that 81% of respondents say they’re either investing in updating AI/ML infrastructure or have definite plans for it.
Moreover, 40% already have a center of excellence to support the works of their organizations’ business units. Yet many improvement opportunities are going unrealized.
Only a quarter reported having AI models fully deployed. Further, 57% said that their organizations’ AI proof-of-concepts are not delivering expected business outcomes. An AppDev and operations director at an enterprise pharmaceutical firm confirmed this insight: “About a third of our AI projects were successful. AI is hard to test and implement. Something can work in theory, but if data is not in the right condition, not clean, not accurate, you can’t train the model.”
#2. Improving AI outcomes requires solving vital data challenges.
Available data has exploded in every enterprise. And AI professionals want this data. The study found that 80% of AI/ML experts exhibited a strong interest in sourcing and using unstructured and edge data from sensors and physical process. However, without the right data management capabilities and critical data infrastructure, organizations underutilize this valuable asset despite having sophisticated AI resources.
For example, 76% of HPC and AI/ML experts thought it was important to reduce analytic silos to improve business decision-making ability and drive insights.
In addition, two-thirds of all respondents found untapped potential in improving the scalability and agility of the data environment so their organizations can quickly evolve and adapt.
#3. AI and HPC together, supported by big data technology, is the future.
Experts from the study envisioned numerous high-impact use cases of AI, HPC, and big data combined. Together, they said these technologies solve for many data challenges with large unstructured data sets, leading to opportunities for AI and HPC to help each other.
The study showed firms pursuing this in financial modeling, simulation, process automation, and risk analysis. Specifically, two-thirds of experts expected cost reduction benefits from using AI to optimize HPC workflows, increasing model targeting accuracy that enabled firms to run fewer simulations.
Sixty-five percent expected higher competitiveness with faster insights when adding HPC to accelerate ML model training. Sixty-one percent believed that AI and HPC together drive the ability to extract newer, unprecedented insights.
A senior strategic sourcing manager at an enterprise manufacturer stated: “I’m anticipating synergy between AI and HPC in the next three to five years in the form of infrastructure convergence. That would enable us to expand the platforms to new users, democratize the data more. In theory, there should be speed as well.”
What is your organization doing to leapfrog toward advanced analytics?
To learn more:
Read the commissioned study by Forrester Consulting: AI Plus HPC: The Future of Advanced Analytics
Visit the Discover More Network to hear the latest advancements in enabling HPC and AI at HPE.
Brandon Draeger
Hewlett Packard Enterprise
twitter.com/brandondraeger
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/info/hpc
Brandon_Draeger
Brandon leads the Compute Product Marketing teams for HPE and joined the company in January 2020 as part of the Cray acquisition. Prior to Cray, Brandon held leadership roles in engineering, product management, marketing, and strategy at Intel, Dell, and Symantec.
