Get faster analytics at a lower cost with GPU acceleration in this private technical preview.
Overview
Read the announcement
The new economics of analytics
AI is increasing the demand for faster, more cost-efficient access to trusted data. As data volumes continue to grow, organizations are under pressure to deliver insights in real time while controlling infrastructure costs. GPU-accelerated query processing helps improve price-performance without moving data or rewriting applications.
Features
How it works
Accelerate supported analytics across data stored in multiple platforms through open table formats—without moving data or rewriting applications.
Supported SQL operations are accelerated by offloading processing from the CPU to NVIDIA GPUs, reducing execution times for large-scale analytical workloads.
Shorter query runtimes deliver faster answers without significantly increasing infrastructure costs.
Provide fast, scalable access to trusted enterprise data so more AI applications and agents can query and act simultaneously.
Case studies
Real-World Impact with Nestlé
“Working with IBM and NVIDIA, early results show we can refresh global operations data in minutes at reduced cost – making a capability that can be turned into tangible business impact in areas such as manufacturing or warehousing.”
Chris Wright,
Chief Information and Digital Officer of Nestlé
83%
Cost Savings – Comparing CPU vs GPU performance
30x
price-performance improvement - Order-to-Cash Data Mart – Updates 3 min vs 15 min
Private Technical Preview
What you receive
A dedicated preview environment
Deployed in your AWS account with governance and access controls defined by your team.
Direct access to IBM experts
A dedicated client advocate, engineering support and a shared Slack channel throughout the evaluation.
Workload-specific results
A final performance and cost assessment comparing GPU-accelerated execution with your current approach.
What participation involves
A focused 2–4 week evaluation
Participate in onboarding, workload testing, progress reviews and a closing results session.
Representative queries and data
Provide anonymized queries and representative data that reflect a meaningful analytical workload.
Candid product feedback
Share what worked, what did not and what capabilities would make the solution more valuable.