Flexible CPU Scaling

Consolidation isn't just possible, it’s now practical.​ With VAST you can scale CPUs independently of the capacity that store data in a VAST cluster — only buy performance when need more IOPS.


DASE Enables Composable QoS

Most commercial storage architectures require clusters to be built from homogenous pools of servers and storage. So when it comes time to add performance, most storage administrators are faced with very rigid rules for adding new nodes to a cluster or even worse, forklift upgrades of whole clusters.

The fully disaggregated architecture of the VAST Data Platform is designed to support asymmetrical expansion, making it simple to add heterogeneous resources to the shared pools of CPU resources and SSDs that compose a VAST cluster. With compute decoupled from storage, adding new CPUs to a VAST cluster, there is no need for a lengthy expansion process that impacts the performance typical of legacy architectures.

How It Works

Flexibility: Powered by DASE™

The path to asymmetry is paved with a few architectural advantages that VAST enjoys, namely that flash storage breaks a long-standing tradeoff between performance and capacity that has been exhibited by HDDs. Unlike HDDs which long ago shed any illusions that performance is increasing across HDD generations, the long-term evolution of flash has shown that performance (in terms of IOPS and bandwidth) has evolved proportionately with capacity. ​


Avoid shared-nothing storage systems' hard-coded performance personalities.

VAST’s new Disaggregated and Shared-Everything (DASE) architecture breaks the tight coupling of compute and storage. By leveraging next-generation low-latency NVMe SSD and networking technology, it’s now possible to disaggregate CPUs from the state of the cluster and for each stateless containerized storage server to simultaneously access all of the data using shared-everything concepts.