3
times faster average training*
2.5
petaFLOPs of AI Performance
2
Cables to Start up
0
Data Centres
A Table-top AI Data Centre in a Box
Dependable & Reliable AI
In Robovision’s field of expertise – agriculture, manufacturing and healthcare – it’s the data scientists who are innovating to transform the world: state-of-the-art robovics, life-saving medical treatments and better-quality products.
Developing such innovation needs a powerful, dedicated and reliable platform on which innovative companies can depend. The NVIDIA DGX Station A100 increases the time-to-value and reduces the CAPEX investment of bringing AI-powered infrastructure and capabilities to your organisation without ever needing a data centre.
Fully-optimised & Ready for AI
The NVIDIA A100 DGX brings data centre-grade performance of 2.5 petaFLOPS for your team to conduct AI training, inference and analytics all on one system. The DGX Station A100 is the only system in existence with four dedicated NVIDIA A100 tensor Core GPUs with multi-instance GPU (MIG))-capabilities. They have the power to deliver the equivalent power of 6 28 dual-core CPU servers each.
The DGX Station A100 is designed with built-in 1Gbase-T baseboard remote management to serve multiple users in multiple locations simultaneously. What’s more is that the NVIDIA has expertly designed the DGX Station A100 to be fully-optimised across hardware and software all powered from a standard electrical outlet.
3 Benefits of Owning a NVIDIA DGX Station A100
1. Reduce CapEx to Enable AI
The NVIDIA DGX Station A100 is especially designed to remove the burden of setting up AI and brings immediate data-centre grade performance to your organisation in a matter of hours. Traditional infrastructure silos are constrained and it is very likely that your organisation will build or rent excess capacity above your immediate needs. Not only this but the data centre maintenance and rental fees can run into the hundreds of thousands per month reducing your organisation’s bottom line.
2. Flexible & Agile AI Infrastructure
With a form factor of no more than an average tower computer, the NVIDIA DGX Station A100 is ideal to be a flexible and agile solution for your AI needs. The NVIDIA DGX Station A100 has a height of 64 cm (25.1 in) and a weight 25 cm (10 in) making it possible to install in the research lab, corporate headquarters or even to enable remote-working. It’s powered by a standard electrical outlet and produces less than 37 decibels (db) of noise – as quiet as having a walk in a remote country location.
3. Greater Analysis & Better Results
With a NVIDIA DGX Station A100, your company’s data science team has access to all the processing power needed immediately and into the future. It is built of AI and is capable of running training, inference and data analytics workloads simultaneously thanks to its multi-instance GPU (MIG) capabilities. The DGX Station A100 with MIG provides up to 28 separate GPU devices to distribute to individual users and projects across the system.
NVIDIA DGX Station A100 Specifications
- Performance: 2.5 petaFLOPS AI 5 petaOPS INT8
- Power Usage: 1.5 kW at 100–120 Vac
- CPU: Single AMD 7742, 64 cores, 2.25 GHz (base)–3.4 GHz (max boost)
- System Memory: 512 GB DDR4
- DGX Display Adapter: 4 GB GPU memory, 4x Mini DisplayPort
- Packaged System Weight: 57.93 kg (127.7 lbs.)
- Dimensions: 639 x 256 x 518 mm (25.1 x 10.1 x 20.4 in)
- Operating Temperature Range: 5–35 ºC (41–95 ºF)
Unrivalled AI Experience & Expertise
Robovision has been a partner of NVIDIA since 2017. Our AI expertise and knowledge has been associated specifically in agri-food, healthcare and manufacturing industries. Robovision has a dedicated team of trained and certified NVIDIA DGXperts to give guidance and assistance to help you become AI-enabled with a NVIDIA DGX Station A100.
Disclaimer
All facts and figures taken from official NVIDIA sources provided to Robovision as a “NVIDIA Preferred Partner”. NVIDIA, the NVIDIA logo, DGX, DGX A100, DGX POD, DGX Station, DGX SuperPOD, and NVLink are trademarks and or or registered trademarks of NVIDIA Corporation. Features, pricing, availability, and specifications are all subject to change without notice by NVIDIA.
All content has been proofread and approved by NVIDIA. 12 April 2022