MayaNAS · parallel file & NAS · on object storage

Feed GPUs from object storage.

MayaNAS serves Lustre parallel file and familiar NAS — NFS, pNFS, SMB — directly out of your cloud's object storage. Training data streams at GPU speed, checkpoints land at tens of gigabytes per second, and capacity costs what your buckets cost. Validated by MLPerf Storage benchmarks.

See it on your cloud → Read the docs one terraform deploy · AWS · GCP · Azure
13 GB/s
sustained from Google Cloud Storage, one filer
11.5 GB/s
to a single NFS client
MLPerf
validated by MLPerf Storage benchmarks
S3 · GCS · Blob
your buckets are the capacity tier
What you get

The filesystems your AI stack already speaks.

Lustre parallel file

One namespace, hundreds of compute clients, striped throughput that scales with the cluster — the same parallel filesystem the world's largest GPU clusters train on, with standard clients.

Cloud-native Lustre in depth →

Familiar NAS alongside

NFS, pNFS, and SMB from the same system for everything around training — notebooks, tools, home directories, render farms. Full POSIX, nothing to install.

pNFS with MayaNAS data servers →

Direct on object storage

Object storage is the substrate, not a backup target — data is read and written in place. No staging pipeline, no cache tier to size and warm, and capacity grows at bucket prices.

Built for checkpoints

Absorbs the burst writes of LLM checkpointing at tens of gigabytes per second, then drains to object storage — training resumes fast, and every checkpoint lands durably.

Highly available

Active-active server pairs with automatic failover. Clients reconnect and carry on — no rerun of the training job, no manual recovery steps.

Terraform-deployed

Deploy from your cloud marketplace into your own account and VPC. Sized configurations from entry to ultra, ready in minutes, billed through your cloud agreement.

Self-serve

Run it yourself.

Deploy with terraform into your own account and VPC — sized configurations, ready in minutes, billed through the cloud agreement you already have. Prefer a guided first run? Talk to us.

AWS

Terraform deployment into your AWS account — S3 as the object tier.

Google Cloud

Terraform deployment into your GCP project — GCS as the object tier.

Azure

Terraform deployment into your Azure subscription — Blob as the object tier.

zettalane-terraform — the modules, sized configurations, and step-by-step guides for every cloud, in one repository.

Parallel file performance, at object-storage cost.

Self-serve with terraform from your cloud marketplace — ready for your most demanding training runs.