--- title: "VAST DataStore" slug: "dev-datastore" updated: 2026-10-28T02:15:22Z published: 2026-10-28T02:15:22Z canonical: "docs.vastdata.com/dev-datastore" --- > ## Documentation Index > Fetch the complete documentation index at: https://docs.vastdata.com/llms.txt > Use this file to discover all available pages before exploring further. # VAST DataStore VAST DataStore is the file and object storage of a VAST cluster. Start by asking your platform team for a bucket and key pair, as Access describes, then run the Quickstart to write and read an object with boto3. Read Concepts before you design an application, and Multiprotocol access before you share data between S3 and NFS. Then go to the guide for your task and check S3 compatibility for the calls you use. ## Overview and concepts Where to begin: the overview and quickstart, then the concepts and architecture the rest of the section relies on, and the release notes. - **[Overview](/develop/docs/dev-datastore-overview)**: VAST DataStore is the cluster's file and object storage over S3, NFS and SMB, for developers building against VAST S3 and NFS. - **[Quickstart](/develop/docs/dev-datastore-quickstart)**: Connect to VAST S3 with boto3, then write an object with metadata, list it, read it back and delete it. - **[Concepts](/develop/docs/dev-datastore-concepts-index)**: How buckets relate to views, who the cluster thinks you are, what allows a request, and what happens after you write. ## Guides Step-by-step instructions for the work applications do on VAST S3 and NFS. Start with Work with S3 objects; read Multiprotocol access before S3 and NFS share data, and the other guides for large objects, versioning, NFS applications and performance. - **[Work with S3 objects](/develop/docs/dev-datastore-guides-s3-objects)**: Write objects with metadata and tags, list by prefix, read ranges, copy and rename, make conditional requests and share presigned URLs. - **[Large objects and multipart uploads](/develop/docs/dev-datastore-guides-large-objects-and-multipart)**: Upload and download large objects with boto3 managed transfers or the multipart calls, understand multipart ETags, and clean up abandoned uploads. - **[Versioning and object lock](/develop/docs/dev-datastore-guides-versioning-and-object-lock)**: List, restore, undelete and permanently delete object versions, protect them with legal holds and retention, and see what differs from AWS. - **[Multiprotocol access](/develop/docs/dev-datastore-guides-multiprotocol)**: Map S3 keys to NFS paths on one view, share data across both protocols, and see what decides ownership and permissions. - **[NFS for applications](/develop/docs/dev-datastore-guides-nfs-for-applications)**: Choose an NFS version and mount options, then publish files atomically, lock byte ranges, use extended attributes and understand client caching. - **[Performance](/develop/docs/dev-datastore-guides-performance)**: Spread parallel S3 and NFS connections across the cluster's virtual IPs and size transfers in boto3 and on NFS mounts for throughput. ## Integrations Settings for the clients and tools that reach VAST S3. Start with Python clients for boto3, s3fs and PyArrow; use Spark and Hadoop for S3A jobs, and AWS CLI, s3cmd and rclone for the command line. - **[Python clients](/develop/docs/dev-datastore-integrations-python)**: Configure boto3, s3fs (for pandas and Dask) and PyArrow's S3FileSystem for a VAST S3 endpoint, including HTTPS certificates. - **[Spark and Hadoop](/develop/docs/dev-datastore-integrations-spark-and-hadoop)**: Configure Hadoop's S3A connector for a VAST endpoint, tune its throughput, and commit Spark job output safely without atomic renames. - **[AWS CLI, s3cmd and rclone](/develop/docs/dev-datastore-integrations-aws-cli-and-tools)**: The endpoint configuration that AWS CLI, s3cmd and rclone each need to move data to and from VAST S3. ## Reference What VAST Cluster 5.5 supports and where it differs from AWS. Start with S3 compatibility and Limits; the NFS reference and SMB reference cover the file protocols. - **[S3 compatibility](/develop/docs/dev-datastore-reference-s3-compatibility)**: Supported S3 requests and headers on VAST Cluster 5.5, the differences from AWS S3 that affect applications, and VAST additions. - **[Limits](/develop/docs/dev-datastore-reference-limits)**: Size and count limits for bucket names, object keys, parts, tags, NFS file names and VAST Catalog on VAST Cluster 5.5. - **[NFS reference](/develop/docs/dev-datastore-reference-nfs)**: Which NFS versions and features VAST Cluster 5.5 supports for clients, plus the access rules and name rules that affect them. - **[SMB reference](/develop/docs/dev-datastore-reference-smb)**: How SMB 2.1 and 3.0 meet S3 and NFS on the same data: Active Directory, shared identities and security flavors. ## Operations What to ask your platform team for, and how to automate the storage an application needs. Start with Access; use CI/CD and automation when a pipeline creates buckets, users and policies. - **[Access](/develop/docs/dev-datastore-operations-access)**: Request a bucket, key pair and policies from your platform team, use temporary credentials, and check what your key can do. - **[CI/CD and automation](/develop/docs/dev-datastore-operations-ci-cd)**: Create a bucket's view, user, S3 key and identity policy from CI with vastpy, the REST API or Terraform. ## Troubleshooting Symptoms, causes and fixes for S3 and NFS problems. Start with Troubleshooting VAST DataStore and find the error or behavior you're seeing. - **[Troubleshooting VAST DataStore](/develop/docs/dev-datastore-troubleshooting-index)**: Causes and fixes for connection, certificate and access-denied errors, object requests, multiprotocol data and NFS problems, by symptom.