NetApp to Acquire PEAK:AIO to Boost Scalable AI Storage
NetApp plans to acquire PEAK:AIO, a parallel file system pioneer, to accelerate AI infrastructure as GPU clusters demand new storage architecture.
NetApp (NASDAQ: NTAP), the San Jose-based intelligent data infrastructure company, has announced its intention to acquire PEAK:AIO, a specialist in next-generation metadata architecture and high-performance parallel file systems. The deal signals NetApp's recognition that the AI era is placing fundamentally different demands on enterprise storage than anything traditional architectures were built to handle.
The strategic rationale is rooted in a structural problem reshaping data centers: as AI workloads proliferate and GPU clusters expand, conventional storage systems are struggling to keep pace with the scale, concurrency, and throughput requirements of modern AI factories and AI cloud environments. PEAK:AIO's parallel namespace technology is specifically engineered to let shared storage scale in lockstep with growing GPU deployments — a capability that has become a competitive prerequisite rather than a differentiator.
Read more Hyperscale Data Eyes 590 MW AI Data Center Expansion in Michigan and Montana →
Central to the acquisition is PEAK:AIO's approach to separating metadata services from raw data, enabling each layer to scale independently. NetApp has been building toward an architecture that can manage metadata at what the company describes as a practically unbounded scale — a critical design goal as AI-driven applications generate file counts and access patterns that overwhelm legacy systems. By folding PEAK:AIO's technology into its roadmap, NetApp aims to close that gap faster than organic development would allow.
The move reflects a broader consolidation trend in enterprise AI infrastructure, where storage vendors are under pressure to demonstrate that their platforms can underpin large-scale model training and inference pipelines without becoming a bottleneck. For NetApp, acquiring specialized parallel file system expertise rather than building it in-house suggests urgency — the window to establish credibility as a native AI infrastructure provider is narrowing as hyperscalers and startups alike compete for the same enterprise budgets.
Continue reading at BusinessWire.
