Get in touch with our cloud team.
NVIDIA DGX Cloud
Train AI models without investing in server hardware
AI Training as a service by NVIDIA
NVIDIA DGX Cloud reduces the cost and complexity of training AI models. Rather than investing in onsite hardware, you rent processing time on systems hosted by NVIDIA, simply paying for what you need, when you need it.
Designed for training and inferencing complex AI models NVIDIA DGX Cloud delivers outstanding performance in LLMs and generative AI.
CONTACT US TO DISCUSS YOUR REQUIREMENTS >
Accelerated AI with NVIDIA DGX
Featuring the most powerful NVIDIA GPUs, high-speed networking and storage, NVIDIA DGX appliances are the gold standard for AI model training and inferencing. The smallest DGX Cloud instance includes four DGX appliances, scalable to as many as your projects need.
| NVIDIA DGX appliances available in DGX Cloud | |||||||
|---|---|---|---|---|---|---|---|
| MODEL | DGX B200 | DGX H100 | DGX A100 | ||||
| GPUS | 8x NVIDIA B200 | 8x NVIDIA H100 | 8x NVIDIA A100 | ||||
| GPU RAM | 180GB PER GPU, 1.44TB TOTAL | 80GB PER GPU, 640GB TOTAL | 80GB PER GPU, 640GB TOTAL | ||||
| CPUS | 2x INTEL XEON 8570 | 2x INTEL XEON 8480C | 2X AMD EPYC 7742 | ||||
| SYSTEM RAM | 4TB | 2TB | 2TB | ||||
| STORAGE | 10TB | 10TB | 10TB | ||||
NVIDIA Base Command
NVIDIA DGX Cloud is managed and controlled by NVIDIA Base Command software. It includes enterprise-grade orchestration and cluster management, libraries that accelerate compute, storage and network infrastructure and system software optimised for running AI workloads.
Trusted By Nvidia
The same software that supports NVIDIA’s thousands of in-house developers, researchers, and AI practitioners underpins every BasePOD.
Scheduling and Orchestration
Base Command provides Kubernetes, Slurm, and Jupyter Notebook environments for DGX systems, delivering an easy-to-use scheduling and orchestration solution.
Comprehensive Cluster Management
Full-featured cluster management automates the end-to-end administration of DGX systems - whether it’s initial provisioning, operating system and firmware updates, or real-time monitoring.
Optimised for DGX
The Base Command software stack is optimised for DGX BasePOD environments that scale from two nodes to 40 node clusters, ensuring maximum performance and productivity.
NVIDIA AI Enterprise
NVIDIA AI Enterprise accelerates the development and deployment of AI projects. Rather than trying to assemble thousands of co-dependent libraries and APIs from different authors when building your own AI applications, NVIDIA AI Enterprise removes this pain point by providing the full AI software stack including applications such as healthcare, computer vision, speech and generative AI.
Improved Productivity
By providing pre-built applications NVIDIA AI Enterprise frees up your team to build innovative AI solutions without having to manage thousands of disparate, open source software.
Enterprise Support
Enterprise-grade support is provided for NVIDIA-certified systems with direct access to NVIDIA’s AI experts, minimising risk and downtime, while maximising system efficiency and productivity.
Building Generative AI
NVIDIA AI Enterprise also includes frameworks such as NeMo, BioNemo and Picasso to jumpstart building generative AI applications.
Start your cloud journey
Scan Cloud workstations are available in increments as small as one month and can be customised to your requirements.
FAQs
How do I get started with Scan Cloud?
Get in touch with our experts to discuss your project goals using live chat, contact form or telephone 01204 474747.