Recognizing that customers don’t want to keep old AI data on its high-speed parallel file system storage, VDURA has partnered with Wasabi to have its low-cost cloud storage available to VDURA customers.
VDURA customers could use Amazon cloud storage, of course, but then they'd have to pay egress fees and be subject to API charges as well. If they use Wasabi, they don't get these charges and pricing is predictable. The two companies are partnering to help AI factories, Neoclouds, and enterprise HPC environments to keep active datasets on VDURA’s GPU-adjacent storage and move old data to Wasabi's public cloud storage through its S3 interface. Datasets, checkpoints, model versions and derived AI artifacts that are no longer in active use move to Wasabi for protection, compliance, disaster recovery and distribution to other sites or compute environments. They can be moved back for retraining, model comparison and governance.
VDURA CEO Ken Claffey said: “The infrastructure that feeds GPUs is engineered for velocity, and data belongs there while it is doing active work. It should not live there permanently. Our customers generate enormous volumes of checkpoints, dataset versions and model artifacts that hold real long-term value. Pairing VDURA's GPU-adjacent performance with Wasabi's open, predictable cloud layer gives them a place for that data to be retained, protected and reused, and makes data movement a native part of the AI data lifecycle.”
Cloud storage provider Backblaze, has started offering its cloud storage facilities to similar customers, with its marketing led by a $335 million deal with Neocloud leader CoreWeave.
All flash storage is becoming very expensive due to NAND production constraints, increased demand from AI, and inability to add more manufacturing capacity for several years. It makes obvious sense to only keep active data on high-speed storage next to GPUs and move old data off to cheap disk storage in the public cloud.
Wasabi has done a similar deal with VAST Data customer Megaport. Laurie Mitchell, SVP Global Marketing, Wasabi Technologies, hammered the point home, saying: “AI data does not lose its value when a training run ends. It becomes the raw material for the next model, the audit trail for governance and the baseline for comparison. This alliance makes it straightforward to connect GPU-adjacent infrastructure with independent, S3-compatible cloud object storage, so organizations keep control of their data and their costs without locking either one into a hyperscaler.”
We may well find similar deals set up by high-speed storage suppliers wanting to offload aging AI data storage to lower-cost public cloud storage or even to tape, with Versity offering fast S3 object access though its gateway.
The VDURA-Wasabi technology alliance will be present at Ai4 2026, August 4 to 6 at The Venetian in Las Vegas, where Wasabi will exhibit at booth #935.