The VAST AI Operating System is a software platform that simplifies and accelerates the entire AI pipeline - from raw data ingestion, through context analysis, to intelligent action - as a continuous, real-time flow. It provides a unified data platform for AI, analytics, data-intensive applications, events, and high-performance storage. AI OS brings together file, object, block, and database services, including hyperscale vector, serverless functions, and other data services within a common architecture that supports a wide range of AI workloads.
Use the guides below for deployment, configuration, administration, monitoring, and maintenance information for each AI OS release.
The VAST Management System GUI (VMS GUI) comprises various sections that enable you to set up networking, manage users and providers, control data access, and monitor performance.
VMS is the control plane for administrators to manage tenants, views, access, quotas, and system configuration.
It supports GUI and REST API access.
Each tenant can be restricted to only see their own resources.
Access can be limited to specific source IPs.
The universal storage foundation for the VAST AI OS. The VAST DataStore provides persistence, protects data via erasure coding, reduces data, and manages the media in a VAST cluster, storing files, objects, volumes, and tables as native Elements in a single namespace: the VAST Element Store. The VAST DataStore presents files, objects, and volumes via standard access protocols, and VAST Tables to the VAST DataBase.
The VAST DataSpace is a blanket term for the set of features that share/move data between VAST clusters, including VAST on Cloud clusters. These include: VAST Global Access, also known as VAST Global Namespace (note that DataSpace was in the past used to mean Global Namespace only, but the definition was expanded); replication; and both synchronous and asynchronous Global Clones
VAST Database delivers ACID compliance for mission-critical operations while enabling instant analytics on the same data—all with zero maintenance.
The VAST Database combines transactional integrity with analytical scale, supporting fast updates and complex queries on the same platform. Newly written data lands instantly in persistent memory, then converts into optimized columnar chunks for high-speed analytics. Unlike traditional architectures that force a tradeoff between real-time operations and large-scale analysis, VAST processes data continuously—so insights, dashboards, and applications always reflect the most current information.
VAST Event Broker: Kafka API-compatible event bus; may also be considered part of DataEngine.
VAST DataEngine is the VAST compute orchestration framework. The framework enables users to write, deploy, and manage execution pipelines triggered by scheduled or event-driven events. VAST DataEngine runs on the VAST Cluster and dynamically provisions infrastructure and runtime resources based on demand. VAST DataEngine frees users from the need to manage infrastructure scaling and availability, or to worry about performance.
Tenants represent isolated administrative and data domains within the VAST cluster.