Domino Data Lab and NetApp demonstrate how Amazon FSx for NetApp ONTAP enables regulated AI operations through governed data architecture, reproducible states, and efficient scaling.

Versions of NetApp StorageGRID 11.3 and later can be tricked into accepting commands from someone who knows how the system works and can reach it on.

Latest acquisition sees firm behind Los Alamos National Lab data project join up

NetApp wants to acquire Peak:AIO to provide the kind of orchestration needed to build high-performance exascale data infrastructures targeting AI workloads.

Reported Monday, Nov 2 - 8:16am EST (TheFly subscription required to read more.)

AI growth in regulated environments put pressure on the systems that govern your data. You need to give teams timely access to approved datasets without losing control of data placement, lineage, or reproducibility. In this post, we share three data architecture principles, show how Domino Data Lab applies them with Amazon FSx for NetApp ONTAP, and give you workload-level metrics to evaluate the architecture.

I measured six paths to the data on Amazon FSx for NetApp ONTAP. Four reimplement the same file-processing job in .NET / Java / Python and run it over NFS, S3 Access Points, and ONTAP REST API so they can be compared. The other two (Trident on EKS, and Athena plus Bedrock Knowledge Bases) are different workloads verified in other repositories. The data does not move; only the access path is added, and I watch how authentication, authorization, and failures change.

AWS DataSync Enhanced mode now supports FSx for Windows File Server, FSx for NetApp ONTAP, and Red Hat OpenShift Virtualization for expanded data transfer capabilities.