ACQUISITION NFS AI Sep 25, 2026
Blocks & Files reported (2026-09-25) that NetApp is acquiring Peak:AIO, a software-defined all-flash storage company founded in 2021 in Manchester, UK. Terms were not disclosed; the transaction remains subject to customary closing conditions and regulatory approvals. Blocks & Files notes the deal is likely not material in a financial sense to NetApp, which reported $410M in free cash flow in its latest quarterly results, and that Peak:AIO secured around $7M in seed funding in October 2025.
Peak:AIO’s software is described as providing a global file namespace and parallel NFS (pNFS) data access for massively parallel workloads; the report says the technology originated from collaborations with research institutions including Los Alamos National Laboratory and Carnegie Mellon University. The company is led by founder and CTO Mark Klarzynski and co-founder Eyal Lemberger (Chief of AI and Cyber Security Strategy), with US-based CEO Roger Cummings joining in May 2024.
What NetApp says the architecture buys. NetApp frames the deal as accelerating its AI infrastructure roadmap by building an architecture that disaggregates metadata from data and lets metadata services scale independently — aimed at environments with trillions of files, exabyte-scale capacity, and massively parallel workloads. NetApp states that ONTAP with added Peak:AIO metadata capabilities and parallel namespace features will preserve ONTAP’s resilience, security, and operational maturity, and that there will be a clear evolution path for existing ONTAP customers.
CEO George Kurian: “AI clouds need high-performance shared storage that can scale alongside growing GPU clusters while maintaining the resilience, security, and operational simplicity organizations depend on. With PEAK:AIO, we are delivering on our vision for a new generation of AI infrastructure that combines scalable metadata services with the proven foundation of ONTAP to help customers maximize infrastructure efficiency and support AI at hyperscale cloud.”
Why it matters to ONTAP operators: parallel NFS is not a new concept in ONTAP — NFSv4.1 already includes pNFS support, with FlexCache acting as the layout server (see the NFS reference). What this acquisition signals is a GPU-era push toward metadata scale-out rather than merely more throughput per controller: the bottleneck for AI factories is often file and metadata operations, not raw bandwidth. Treat the announcement as directional — the productised ONTAP integration is not shipped or dated in the sources, so do not plan refreshes around it yet; the deal itself is still subject to closing conditions. The report’s bootnote adds context on Peak:AIO’s hardware: a 2RU parallel-NFS DataServer based on a Dell server with Solidigm NVMe SSDs and NVIDIA CX7 Ethernet NICs, plus an NVMe-flash PEAK:ARCHIVE product offering up to 1.4 PB in 2RU.
Read the original (Blocks & Files) · NFS reference (pNFS, FlexCache layout server) · Protocol choice guide · AI coverage index