VAST DataStore

Unconventional Thinking​ for Unstructured Data​

Meet the revolutionary VAST DataStore, engineered in the age of AI.​

Meet the revolutionary VAST DataStore.​ A universal storage platform engineered in the age of AI.​

Read the White Paper
Features
Resources
Trusted by the world’s leading data-driven organizations
View All Customers

No more band-aids, compromises, or tiers.

VAST Data was founded on the idea that the future of artificial intelligence must be built upon fast infrastructure that allows for AI engines to process data at any scale.​

To fulfill this vision, we’ve engineered the first new distributed system architecture in 20 years with the mission to eliminate compromise and unify storage environments into a single scalable, affordable and resilient tier of flash.​

Once your archive is built from flash, you no longer need additional tiers of infrastructure above it.  No more hard drives. No more tiers. Storage is now just universal and… simple.​

images

Purpose built for all your data.

Imagine the possibilities if you had the freedom to use your data anyway you wanted. If your application could instantly access all the data it needs from a pool of all the images, files, and data you own. If you could linearly scale performance and capacity as needed from petabytes to exabytes at the cost of an archive.

The VAST Data story started with the DataStore: an enterprise network attached storage platform built to meet the needs of today’s powerful AI computing architectures and beyond.

Industry

Use Case

Technology

VAST Data Platform

Redefining Flash Economics​

DASE efficiency algorithms bring the cost of flash in-line with HDD storage. Similarity delivers world-leading data reduction. New Erasure Codes minimize the overhead of failure protection. New QLC flash management protects QLC for a decade.​

Edge-to-Cloud Global Namespace​

The VAST DataSpace is designed to provide global data access across all the locations you compute. A decentralized approach to distributed locking makes it possible to transact with performance and consistency at the edge.​

Simple & Scalable Snapshots​

We’ve pioneered new data structures that make it possible to easily snapshot data even at exabyte scale. VAST Snapshots never require data copies, so performance is never impacted. Easily take and replicate 100,000s of snaps at any system depth.​

Eliminate Refreshes & Migrations​

VAST’s asymmetric cluster architecture can combine multiple generations of flash and CPU hardware into a single cluster without exposing tiering complexity to users. This approach eliminates hardware refresh cycles and painful data migrations.​

Flexible Quality-of-Service​

The DataStore features multiple levels of QOS to enable scalable service delivery and infrastructure consolidation. Logical and HW-based QOS tools make it possible to either govern or isolate I/O flows from noisy neighbors or whole compute clusters.​

Better for the Environment​

VAST-powered systems are simply better for the environment. VAST’s efficiency algorithms provide absolute storage-level power savings, and by enabling real-time I/O even down to the archive…VAST systems save on compute power too.​

Redefining Flash Economics​

DASE efficiency algorithms bring the cost of flash in-line with HDD storage. Similarity delivers world-leading data reduction. New Erasure Codes minimize the overhead of failure protection. New QLC flash management protects QLC for a decade.​

Edge-to-Cloud Global Namespace​

The VAST DataSpace is designed to provide global data access across all the locations you compute. A decentralized approach to distributed locking makes it possible to transact with performance and consistency at the edge.​

Simple & Scalable Snapshots​

We’ve pioneered new data structures that make it possible to easily snapshot data even at exabyte scale. VAST Snapshots never require data copies, so performance is never impacted. Easily take and replicate 100,000s of snaps at any system depth.​

Eliminate Refreshes & Migrations​

VAST’s asymmetric cluster architecture can combine multiple generations of flash and CPU hardware into a single cluster without exposing tiering complexity to users. This approach eliminates hardware refresh cycles and painful data migrations.​

Flexible Quality-of-Service​

The DataStore features multiple levels of QOS to enable scalable service delivery and infrastructure consolidation. Logical and HW-based QOS tools make it possible to either govern or isolate I/O flows from noisy neighbors or whole compute clusters.​

Better for the Environment​

VAST-powered systems are simply better for the environment. VAST’s efficiency algorithms provide absolute storage-level power savings, and by enabling real-time I/O even down to the archive…VAST systems save on compute power too.​

Simplicity At Scale

High-Performance Client Access​

The VAST DataStore embraces open, standard protocols and marries them with high-performance optimizations to eliminate the need for complex client-side storage drivers like parallel file systems or SAN clients.​

The VAST NFS experience, in particular, takes the classic NAS experience and “turns it up to 11” by featuring support for RDMA, multi-pathing and even Nvidia’s GPUDirect Storage APIs.​

An Enterprise Multi-Protocol Architecture​

All the VAST protocols are written in-house to provide the best performance and permissions experience.  By owning and controlling the full stack, we’ve designed each protocol to work seamlessly with our namespace abstraction we call the Element Store. With VAST, all protocols share equal data access and enjoy equal performance access.​

images

Meet The VAST Catalog: The Semantic Layer for Files & Objects

The VAST DataBase breaks the tradeoff between conventional OLTP systems and data warehouses to provide a transactional and analytical experience that scales without limits.

For unstructured data workloads, the VAST DataStore can record every file and every object with lightning-fast transaction performance into a queryable table that also supports data enrichment using custom user-defined metadata.

The VAST Data Catalog provides users and administrators insight into directories and buckets by enabling fine-grained queries at any scale. We've eliminated the need to perform time-consuming operations that walk file systems, now insights are available in real-time.

Read the Story

A Breakthrough In Unstructured Data Reduction​

Similarity is VAST’s revolutionary approach to global data compression, offering a unique combination of data reduction and fine-grained pattern matching. Similarity goes beyond traditional methods by uncovering correlations in data that others miss to deliver the world’s first exabyte-scale reference compression system.

Similarity can efficiently reduce deduplicated, compressed, and encrypted data, delivering unparalleled savings. VAST guarantees that Similarity will outperform any other solution in the market, promising to cover the difference if it falls short. On average, customers experience 3:1 data reduction:

2:1 Reduction​ Deep Learning & Machine Learning Training Pipelines

2x Smaller Backups​ Vs. Legacy Data ​ Protection Appliances

1.7:1 Reduction​ Animation and Video Post-Processing Pipelines

2.5:1 Reduction​ Multi-Disciplinary HPC Computing Environments

Features

A Next-Generation Architecture Designed for the Enterprise

Multi-Protocol by Design

Eliminate islands of data infrastructure with a single platform built to deliver the integrated data management that can only come from engineering the entire protocol layer in house.

NFS V3, V4

Parallel file system speed with NAS simplicity. VAST accelerated NFS removes the limitations of legacy NFS without the need for proprietary clients so that everyone can support next-gen AI-driven workloads.

SMB 2.1, 3

Support enterprise workloads with multi-channel support for non-disruptive workloads and high performance. VAST's SMB protocol was designed to take full advantage of the DASE architecture so your most demanding applications run faster and smoother than ever.

S3

High-performance object storage for modern data analytics and massive scale workloads.

Container Plugins

Programmatically create persistent and ephemeral K8S volumes. VAST’s Container Storage Interface (CSI) driver provides a standardized interface that can be used by a Kubernetes cluster to provision persistent volumes.

GPUDirect

Achieve predictable, high performance to meet any machine-learning and deep-learning workload requirements with VAST support for NVIDIA's GPUDirect® Storage.

IPv4 & IPv6

Implement next-generation workloads and future-proof your infrastructure with VAST.

Multi-Protocol by Design

Eliminate islands of data infrastructure with a single platform built to deliver the integrated data management that can only come from engineering the entire protocol layer in house.

NFS V3, V4

Parallel file system speed with NAS simplicity. VAST accelerated NFS removes the limitations of legacy NFS without the need for proprietary clients so that everyone can support next-gen AI-driven workloads.

SMB 2.1, 3

Support enterprise workloads with multi-channel support for non-disruptive workloads and high performance. VAST's SMB protocol was designed to take full advantage of the DASE architecture so your most demanding applications run faster and smoother than ever.

S3

High-performance object storage for modern data analytics and massive scale workloads.

Container Plugins

Programmatically create persistent and ephemeral K8S volumes. VAST’s Container Storage Interface (CSI) driver provides a standardized interface that can be used by a Kubernetes cluster to provision persistent volumes.

GPUDirect

Achieve predictable, high performance to meet any machine-learning and deep-learning workload requirements with VAST support for NVIDIA's GPUDirect® Storage.

IPv4 & IPv6

Implement next-generation workloads and future-proof your infrastructure with VAST.

Enterprise by Nature

Enable mission-critical and data-intensive enterprise production environments to consolidate workflows, unlocking fast access to all data.

Exabyte Scale Namespace

Independently scale performance and capacity in a single massive namespace.

LDAP, Active Directory

Secure multi-tenancy with multiple authentication sources.

Multi-Protocol ACL

Multi-protocol access to any data with a unified permission model.

Metadata Index in the VAST Catalog

A built-in, high-speed index of file with point-in-time granularity.

Directory and User Quotas

Optimize your resource utilization with granular quotas.

Snapshots and Object Immutability

Protect data from ransomware and insider threats with indestructible storage.

Similarity-Based Data Reduction

A breakthrough global data reduction algorithm that redefines the economic equation of flash.

Capacity and Performance Analytics

Monitor your workloads via the intuitive VAST UI or export metrics to the log analytics tool of your choice.

Distributed / Multiprotocol File Locking

Ensure consistency of operations across the cluster.

Multi-tenancy

Support multiple securely isolated tennets across a common infrastructure and eliminate resource contention issues.

Quality of Service (QoS)

Create optimized QoS policies for service levels with on the fly flexibility.

S3 Lifecycle Policy

Implement data management policies with cloud workflows.

Access Audit

Monitor your VAST cluster with granular protocol and admin access audit logging.

Enterprise by Nature

Enable mission-critical and data-intensive enterprise production environments to consolidate workflows, unlocking fast access to all data.

Exabyte Scale Namespace

Independently scale performance and capacity in a single massive namespace.

LDAP, Active Directory

Secure multi-tenancy with multiple authentication sources.

Multi-Protocol ACL

Multi-protocol access to any data with a unified permission model.

Metadata Index in the VAST Catalog

A built-in, high-speed index of file with point-in-time granularity.

Directory and User Quotas

Optimize your resource utilization with granular quotas.

Snapshots and Object Immutability

Protect data from ransomware and insider threats with indestructible storage.

Similarity-Based Data Reduction

A breakthrough global data reduction algorithm that redefines the economic equation of flash.

Capacity and Performance Analytics

Monitor your workloads via the intuitive VAST UI or export metrics to the log analytics tool of your choice.

Distributed / Multiprotocol File Locking

Ensure consistency of operations across the cluster.

Multi-tenancy

Support multiple securely isolated tennets across a common infrastructure and eliminate resource contention issues.

Quality of Service (QoS)

Create optimized QoS policies for service levels with on the fly flexibility.

S3 Lifecycle Policy

Implement data management policies with cloud workflows.

Access Audit

Monitor your VAST cluster with granular protocol and admin access audit logging.

Infrastructure Built for Scale & Efficiency

VAST innovations combine to deliver compounded levels of infrastructure efficiencies that make it possible to build an always-on, ultra resilient flash cloud.

Transactional Storage System

Achieve the transactional guarantees of an ACID database with the performance of a parallel file system and the scale of an object store.

Distributed Persistent Metadata

Non-volatile, cluster-wide metadata management for superior resilience, scalability, and performance.

Global Hyperscale Flash Translation

Unleash the potential of ultra-dense, low-cost hyperscale flash for any workload while realizing a decade of SSD durability.

Encryption at Rest and Flight

Protect data in transit and stored on flash with advanced encryption.

Rapid Rebuild Erasure Coding

Revolutionary erasure coding that rebuilds in a fraction of the time of Reed-Solomon codes with as little as 2.7% overhead.

Fail-In-Place Architecture

All controllers fail gracefully to remaining controllers. SSDs rebuild against remaining free space even before failed devices are replaced.

CNode Pooling

Dedicated pools of CNodes provide CPU-based QoS boundaries, serve common data across disparate networks and provide additional security by limiting access on a per-share basis across specific pools.

Asynchronous Replication with Automated Failover

Take control of disaster recovery workflows with automated synchronization and reversed replication.

Cloud Backup

Leverage remote S3 storage for system backups.

Multi-Factor Authentication

Secure administrative access with multi-factor authentication.

Infrastructure Built for Scale & Efficiency

VAST innovations combine to deliver compounded levels of infrastructure efficiencies that make it possible to build an always-on, ultra resilient flash cloud.

Transactional Storage System

Achieve the transactional guarantees of an ACID database with the performance of a parallel file system and the scale of an object store.

Distributed Persistent Metadata

Non-volatile, cluster-wide metadata management for superior resilience, scalability, and performance.

Global Hyperscale Flash Translation

Unleash the potential of ultra-dense, low-cost hyperscale flash for any workload while realizing a decade of SSD durability.

Encryption at Rest and Flight

Protect data in transit and stored on flash with advanced encryption.

Rapid Rebuild Erasure Coding

Revolutionary erasure coding that rebuilds in a fraction of the time of Reed-Solomon codes with as little as 2.7% overhead.

Fail-In-Place Architecture

All controllers fail gracefully to remaining controllers. SSDs rebuild against remaining free space even before failed devices are replaced.

CNode Pooling

Dedicated pools of CNodes provide CPU-based QoS boundaries, serve common data across disparate networks and provide additional security by limiting access on a per-share basis across specific pools.

Asynchronous Replication with Automated Failover

Take control of disaster recovery workflows with automated synchronization and reversed replication.

Cloud Backup

Leverage remote S3 storage for system backups.

Multi-Factor Authentication

Secure administrative access with multi-factor authentication.

Consumption Model

Sold as software, delivered and supported as an appliance.

Meet Gemini – the business of storage, disaggregated. With Gemini, customers purchase managed software on hardware that can now be bought directly from our manufacturers at cost. Gemini provides customers more commercial flexibility and new ways to save on software storage solutions - all while delivering unrivaled levels of scale-out deployment simplicity.

On-Premise, VAST appliances are designed to find the optimized balance of performance and capacity. We collaborate with leading enterprise technology manufacturers to specify resilient, scalable, and efficient equipment. Our scalable Cluster architecture allows for mix and match across generations of flash and storage compute infrastructure.

Learn more about supported platforms.

LightSpeed
Ceres
Supermicro
HPE
Mercury
In partnership with Sanmina, LightSpeed is a 2U HA modern flash enclosure, born in the era of scalable AI. LightSpeed combines the light touch of a scale-out NAS with the speed of a parallel file system to deliver on the promise of simplicity at scale.
View Specs
In partnership with Sanmina, LightSpeed is a 2U HA modern flash enclosure, born in the era of scalable AI. LightSpeed combines the light touch of a scale-out NAS with the speed of a parallel file system to deliver on the promise of simplicity at scale.
View Specs

Trusted by leading companies across industries

Read Reviews
Choosing VAST was viewed as a risk at our organization. It is now considered a big win!
Infrastructure Team Lead in Education
images
Features are being rapidly added and they are being done right from the beginning.
IT Infrastructure Lead at Animation Studio
images
White Paper

NVIDIA DGX SuperPOD: VAST

Reference Architecture - Featuring NVIDIA DGX H100 and DGX A100 Systems

13 pages
White Paper

Big Data Storage Platform Comparison

The shift from Hadoop File System (HDFS) to more scalable and efficient data storage options for modern data analytics requires a balanced platform able to handle diverse workloads.

6 pages