By Leona Lo, Editor-at-Large
At INSIGHT Xtra Singapore 2026 on 10 February, NetApp outlined its data platform strategy aimed at helping ASEAN enterprises move from AI pilots to scaled deployment.
Regional momentum is building. A recent McKinsey–EDB–Tech in Asia report shows 46 percent of Southeast Asian companies are scaling AI initiatives, compared with 35-percent globally. As adoption accelerates, the challenge is shifting from experimentation to operational execution.
NetApp argues that as AI moves from pilots to production, the primary constraint is not compute capacity, but data readiness. While GPU shortages dominate industry discussion, many organisations face bottlenecks in data movement, governance and storage efficiency.
To address this, the company is consolidating data services under its ONTAP platform with the goal of reducing data movement and improving infrastructure efficiency.
Scaling AI in a fragmented regulatory environment
ASEAN enterprises operate across uneven cloud maturity levels and complex data sovereignty requirements. Hybrid environments combining on-premises, private and public cloud infrastructure are now standard.
“2026 is shaping up to be the year of production AI for ASEAN,” said Henry Kho, Area Vice President & General Manager, GCASK (Greater China, ASEAN & South Korea), NetApp. He noted that scaling AI across distributed environments while meeting compliance obligations remains a key challenge.
“As we look to expand our AI capabilities across borders, complying with local data sovereignty laws while maintaining high performance has been our biggest hurdle,” said Louis Xu, Head of Bitdeer AI Cloud. “We need infrastructure that does not just store data but intelligently manages it across our hybrid environment.”
Together, the comments highlight a common issue: applying consistent data control and performance standards across fragmented regulatory landscapes.
Simplifying the generative AI stack
Generative AI deployments often involve separate systems for storage, data lakes, vector databases, GPU clusters and orchestration tools. Managing these components independently can increase latency and operational complexity.
We need infrastructure that does not just store data but intelligently manages it across our hybrid environment.
Louis Xu
NetApp integrates these capabilities within a unified ONTAP architecture built around two core components: AFX and the AI Data Engine.
AFX is a disaggregated all-flash storage system purpose-built for AI. It scales to 128 nodes with terabytes-per-second throughput, supporting large-scale training and inference workloads. Unlike traditional high-availability pairs, AFX separates storage compute from capacity, allowing performance and capacity to scale independently.
The AI Data Engine provides integrated data services within the same environment, reducing the need for external tools.
NetApp reports potential GPU utilisation gains of up to 30% by limiting unnecessary data movement and simplifying the pipeline.
Addressing the Data Readiness gap
Gartner estimates that through 2026, 60-percent of AI projects will be abandoned due to insufficient data readiness.
Beyond infrastructure scale, bottlenecks remain in dataset preparation and version control. Training data must be cleaned, indexed, tested and retrained repeatedly.
NetApp’s AI Data Engine, built on NVIDIA’s AI Data Platform reference design, is designed to streamline this process. Snapshot-based versioning allows faster iteration of training datasets while reducing storage duplication.
By simplifying how datasets are prepared and reused, the platform aims to remove operational delays that often prevent AI initiatives from progressing beyond pilot stages.
As ASEAN organisations move deeper into production AI, infrastructure decisions are increasingly shaped by regulatory complexity, operational efficiency and data lifecycle management. In this environment, data architecture, not just compute capacity, is emerging as a central factor in determining whether AI initiatives succeed at scale.
