Vexata

Vexata

Vexata powers todays cognitive era applications driving predictive analytics and autonomous decisions. Discover how Vexata's Scalable NVMe Flash Storage unlocks the full potential of today's advanced solid-state media to lower the TCO of your data center infrastructure.

Edit ID  12147916 

Edita Rexha
Edita Rexha edited on 18 September, 2019 10:12 pm
Edits made to:
Infobox (+3 properties)
Description (+270 characters)
Article (+1355 characters)
Topic thumbnail

Vexata

Vexata powers todays cognitive era applications driving predictive analytics and autonomous decisions. Discover how Vexata's Scalable NVMe Flash Storage unlocks the full potential of today's advanced solid-state media to lower the TCO of your data center infrastructure.

Article

Cognitive/AI applications operate across large data sets to detect, learn, decide and respond autonomously. Advances in CPU, GPU and FPGA co-processors are dramatically increasing performance while reducing the cost of Cognitive/AI computing. Unfortunately, data performance at scale has not kept pace with these compute advances and is fundamentally limiting application performance.

What's required is a shared data platform that delivers hot tier performance at Cloud scale and economics. Vexata's Accelerated Data Architecture achieves unparalleled throughput at ultra-low latency for the highly parallel, random, mixed read/write data patterns that characterize Cognitive/AI applications. This 3-stage software architecture delivers unmatched performance and efficiency across massive data sets without relying on caching or tiering.

VX-OS, at the core of Vexata's Accelerated Data Architecture consists of 3 primary architectural elements:

...

VX-OS runs as a distributed software architecture within the VX-100 and VX-Cloud hardware subsystems to deliver unmatched performance with ultra-low latency. Together, this powerful, efficient system supports highly parallel, random, mixed R/W data patterns that characterize Cognitive/AI applications. Vexata achieves this transformative performance across large data sets without relying on caching or tiering.

Infobox

Golden logo
Text is available under the Creative Commons Attribution-ShareAlike 4.0; additional terms apply. By using this site, you agree to our Terms & Conditions.