- Integrated Systems
- About Us
- Integrated Systems
- About Us
AI at the edge: What’s the right direction to take?
AI at the edge deployments are often bigger than organizations expect—and quickly turn complex. Having the right expertise is critical to ensure your AI projects get into production and start generating value faster.
We’ve all been there: Torn about which direction to take when making important decisions. Do you make a smart, safer decision? Or go for a riskier one? In reality for most of IT professionals, it’s a little bit of both.
At first glance implementing AI may seem simple. What could go wrong with the compelling value of using data intelligently to derive insights for faster decision making? Nothing right? Wrong. It’s when you start to dig deeper into the details, where the complications become more evident.
Addressing a familiar AI misconception: Infrastructure is the key component
The truth is, not necessarily. While it plays an important role, many tend to overlook other critical factors, such as data, connectivity, and the expertise to setup and run effectively.
In our HPE webinar series, World Watch, we recently talked about enterprise AI from edge to exascale. Arti Garg, head of Advanced AI Solutions and Technologies at HPE, discussed how all these factors combined can help improve proof-of-concept to production success rates with AI at the edge.
With increasing magnitudes of data growth, it’s imperative your datacenter has agile and resilient storage and memory systems that can move immense amounts of data from edge to core at rapid speeds, enabling them to efficiently run AI algorithms and models at scale. But it’s also essential that data and connectivity required can handle this exponential data in a secure, reliable, and accelerated way.
Arti was joined by Ian Hughes from 451 Research, a part of S&P Global Market Intelligence. He shared recent voice of the enterprise survey results, specifically highlighting AI at the edge adoption trends. He also focused on key use cases, including comparisons of where core, near edge, and edge adoptions reside and how private and public cloud options come into play.
Identifying potential obstacles
The reality is, AI workloads are growing larger, more complex, and more diverse than ever. This in turn means risk is higher if not addressed properly. Scaling at the size and velocity expected are causing operational challenges across the enterprise.
As you ramp up your AI initiatives, you need to address common obstacles first.
- Are you solving the wrong problems and scoping AI projects incorrectly?
- Have you secured the necessary funding?
- Are you equating MLOps with DevOps?
- Are your data pipelines ready for real-time data?
- Is your IT infrastructure ready for AI?
In our Operationalizing Enterprise AI: Proven solutions for successful AI deployment paper, experts share what to lookout for as you deal with these obstacles—and how HPE solutions are helping customers realize the value of enterprise AI faster, thanks to proven and practical approaches to create new applications and achieve breakthrough innovations.
AI case study: Making zero defects a reality
Seeing AI at the edge in action makes it all come alive. Relimetrics aids companies who manufacture products by helping them transform how they design and create products using software that fully digitizes their quality audit (QA) cycles. The solution uses real-time GPU-accelerated video analytics and machine learning to inspect the configurations and properties of product components, improving the accuracy of detecting defects.
What results are they realizing?
- Reduced the number of defective products that reach customers by 25%
- Expanded test coverage by 20% and saves 96 seconds of inspection time per server
- Improved overall production performance from sigma 2.1 to sigma 4.
“We are taking QA automation to a new level. Reducing scrap and rework is no longer a dream but a proven reality.” – Kemal Levi, Founder and CEO, Relimetrics
Ready to put AI at the edge into action?
AI at the edge implementations are bigger than you think and turn into complex projects quickly. Having the right expertise is critical to ensure your AI project gets into production and starts generating value for your business faster. Access our AI and data experts who can provide you with personal support and collaboration to plan and execute your ideal solution. An approach many are considering is HPE GreenLake as a service which enables you to free up capital by utilizing a pay-per-use model. Enterprises can revitalize their AI and scale resources as necessary with the agility of the cloud. By centralizing operations and offloading time-consuming management tasks, enterprises can easily execute and adapt new AI use cases with ongoing support and training.
Whether you’re activating fraud detection, video surveillance, personalized medicine, or modeling and simulation, one thing is for certain: AI that is data-driven, production-oriented, and cloud-enabled—available anytime, anywhere, and at any scale—will always be the right decision.
Watch the World Watch webinar on demand: Enterprise AI from Edge to Exascale
Meet HPE Experts blogger Tracy Siclair
Tracy has worked for HPE for 24 years in various positions, all geared toward providing a better customer experience. She has a passion for thinking out of the box and finding innovative ways to share commercial insights. Using research and insights, she’s putting those into action by telling stories customers can easily understand.
Hewlett Packard Enterprise