Universal Storage for Dremio

Lighting Fast Queries Powered by All-Flash Data Lakes

Scalable all-flash is key to ensuring real-time query performance across the whole of your vast data reserves. VAST’s Universal Storage is the only data store for Dremio that makes affordable enterprise flash data lakehouses resilient and simple at any scale.

Trusted by the world’s leading data-driven organizations
View All Customers
View All Customers

Get the power of a high-performance file system, the simplicity of a NAS and the scalability, capability and affordability of a data lake. Universal Storage and Dremio deliver unified data analytics that brings the performance and functionality of a data warehouse to a flash-powered data lake.

Overview

Lightning-fast queries across all of your data.

Today’s data lakehouse requires a combination of real-time storage access, capacity scale and affordability that has never before been available from legacy approaches to all-flash infrastructure. VAST has broken the long-standing tradeoff between performance and the cost of capacity in order to finally make affordable all-flash data lakes a reality for every organization.

Dremio and VAST have partnered to rethink how shared datalake infrastructure can be deployed. Our combined solution delivers the speed of a cache layer all the way across your datalake and enables sub-second query performance at any scale.

images
images

Partnering with VAST ensures Dremio users are equipped with the lakehouse data capacity and scalable high performance necessary to run their business intelligence workloads and data analytics applications. As data volumes continue to grow, VAST’s disaggregated architecture enables users to easily scale the performance and capacity that businesses demand, and that our open data lakehouse platform delivers.

Roger Frey
VP, Alliances, Dremio
Key Benefits

Universal Storage + Dremio: Better Together

For the first time in 20 years, VAST Data has introduced a new type of distributed system architecture. VAST's Disaggregated and Shared Everything (DASE) storage system was built from the ground up to eliminate the tradeoffs imposed by legacy scale-out architectures. Our Universal Storage system was engineered to be fast enough for your most demanding pipelines, scalable enough for all your data and affordable enough that you never need to think about storage tiering again.

Revolutionary All-Flash Economics

Deploy flash across the entire lakehouse with Universal Storage. VAST’s Similarity-Based Data Reduction enables global data reduction that saves space even when dealing with pre-compressed Avro, Parquet and Orc files.

Scale Without Feature Compromise

Get near-limitless snapshots, enterprise replication, user and admin auditing, remote monitoring, online upgrades and expansions and so much more. Break the scalability & data management simplicity compromise.

Flexible, Multi-Protocol Infrastructure

Converge file stores and data lakes while supporting content-rich workloads that use the Dremio Lakehouse. Get real-time access from GPUs to file data for Python workloads powered by NFS, RDMA, and GPUDirect Storage.

A Parallel Data Service Architecture

Scale all-flash clusters far beyond the limits of legacy scale-out systems with DASE, an all-new disaggregated, embarrassingly parallel cluster architecture. Performance scales linearly to support 100s of Dremio Executor nodes.

Best In Class Price-Performance

New approaches to data reduction, flash management, and data protection accelerate all data operations and searches at half the cost of traditional flash solutions, eliminating the need for storage tiering.

No Resource Contention

Dedicated Quality of Service for Dremio’s sub-engines ensures complex data science jobs don’t prevent queries or dashboards from loading. Eliminate nosy neighbors and get predictable performance for select user pools.

Revolutionary All-Flash Economics

Deploy flash across the entire lakehouse with Universal Storage. VAST’s Similarity-Based Data Reduction enables global data reduction that saves space even when dealing with pre-compressed Avro, Parquet and Orc files.

Scale Without Feature Compromise

Get near-limitless snapshots, enterprise replication, user and admin auditing, remote monitoring, online upgrades and expansions and so much more. Break the scalability & data management simplicity compromise.

Flexible, Multi-Protocol Infrastructure

Converge file stores and data lakes while supporting content-rich workloads that use the Dremio Lakehouse. Get real-time access from GPUs to file data for Python workloads powered by NFS, RDMA, and GPUDirect Storage.

A Parallel Data Service Architecture

Scale all-flash clusters far beyond the limits of legacy scale-out systems with DASE, an all-new disaggregated, embarrassingly parallel cluster architecture. Performance scales linearly to support 100s of Dremio Executor nodes.

Best In Class Price-Performance

New approaches to data reduction, flash management, and data protection accelerate all data operations and searches at half the cost of traditional flash solutions, eliminating the need for storage tiering.

No Resource Contention

Dedicated Quality of Service for Dremio’s sub-engines ensures complex data science jobs don’t prevent queries or dashboards from loading. Eliminate nosy neighbors and get predictable performance for select user pools.

Reference Architecture

One universal platform for all data center data.

  • Consolidate

    Run a variety of analytics workloads on the same namespace and eliminate islands of infrastructure

  • Quality of Service

    Resize container pools & eliminate nosy neighbors to tackle workloads of any concurrency

images