Servers in data center
Products > NVIDIA DGX Systems

Power Your Enterprise AI Infrastructure with NVIDIA DGX Systems

Penguin Solutions, a NVIDIA DGX-Ready Managed Services Provider, delivers proven NVIDIA DGX™ systems with rapid deployment, ongoing optimization, and expert support helping you get AI models into production and run workloads at peak efficiency.

Let's Talk
Request Pricing

Penguin Solutions: Your
Certified DGX-Ready
Managed Services Partner

With over 25 years of high-performance computing expertise and hands-on AI experience since 2017, Penguin Solutions helps organizations design, build, deploy, and manage NVIDIA DGX-accelerated AI infrastructure. Our OriginAI® pre-validated architectures, ICE ClusterWare™ cluster management software, and expert services accelerate deployments, streamline operations, and optimize performance.

As a NVIDIA Elite Partner and DGX-Ready Managed Services Provider, we provide expert support across the full AI lifecycle. From planning and deployment to ongoing management, we help organizations fill critical skills gaps, scale confidently, and keep DGX clusters running at peak performance so your teams can focus on AI innovation.

NVIDIA Elite Partner
NVIDIA data center

Accelerate Your AI Initiatives with
Penguin Solutions and NVIDIA

Whether training models or scaling inference, NVIDIA DGX platforms deliver. Penguin Solutions ensures you choose the right system, integrate seamlessly, and optimize your infrastructure's performance.

Latest generation AI systems accelerated by NVIDIA Blackwell architecture delivers unprecedented performance for the most demanding AI workloads including foundation model training and inference.

  • DGX B200: Versatile enterprise AI with 192GB HBM3e memory per GPU for diverse, large-scale workloads
  • DGX B300: Purpose-built for massive-scale generative AI and AI reasoning, with 50% more compute performance (vs. DGX B200) and 288GB HBM3e memory per GPU
  • 72 petaFLOPS of AI performance
  • Advanced NVLink interconnect at 1.8TB/s
  • Optimized for trillion-parameter models
Download Datasheet
NVIDIA DGX B200 & B300

Revolutionary liquid-cooled rack-scale system combining Grace CPUs and Blackwell GPUs. Purpose-designed for the most demanding trillion-parameter AI models and real-time inference at unprecedented scale.

  • Grace-Blackwell Superchip architecture
  • 72 GPU configuration in liquid-cooled rack-scale design
  • GB200 & GB300 NVL options available
  • Unprecedented memory bandwidth and capacity
  • Exascale AI computing in a single rack
Download Datasheet
NVIDIA GB200 & GB300 NVL72

Entry-level AI system designed for organizations starting their AI journey. Delivers enterprise-grade AI capabilities in an accessible, cost-effective platform perfect for development and testing.

  • Right-sized for AI development teams
  • Full NVIDIA AI software stack included
  • Scalable architecture for future growth
  • Simplified deployment and management
Download Datasheet
NVIDIA DGX Spark

High-performance networking infrastructure optimized for AI workloads. Complete portfolio of switches, adapters, and cables delivering ultra-low latency and maximum throughput for AI clusters.

  • Quantum-X800 InfiniBand and Spectrum-X Ethernet
  • Up to 800Gb/s per port connectivity
  • NVIDIA ConnectX and BlueField adapters
  • Advanced congestion control and adaptive routing
  • Integrated with NVIDIA AI Enterprise
Download Datasheet
NVIDIA Spectrum Ethernet Switch Family
ICE ClusterWare™

Intelligent Management for
AI Infrastructure

The Penguin Solutions ICE ClusterWare™ platform helps organizations simplify the deployment and management of AI clusters to realize high productivity quickly. With ICE ClusterWare, bare-metal hardware, network, and software resources are transformed into high-performance cluster environments, streamlining the complexity of administration, and optimizing the availability of computing resources.

ICE ClusterWare helps assure uptime for critical, demanding AI environments by:

  • Delivering rapid provisioning and extensibility
  • Supporting advanced workload schedulers
  • Providing cluster health monitoring and alerting
  • Integrating with industry-standard DevOps tools
  • Supporting the strictest security protocols and restricted environments
ICE ClusterWare on monitor
ICE ClusterWare on monitor
ICE ClusterWare on monitor
Customer Success Story
Meta server racks

Delivering AI-Optimized Architecture and AI Managed Services for Meta

Penguin Solutions worked with Meta’s operations team on the hardware integration to deploy the cluster and set up major parts of the control plane. Penguin’s hardware and software expertise helped to unite contributions from NVIDIA and Pure Storage.

Together, these three partners were key to supplying Meta with an optimized solution—the new AI Research SuperCluster (RSC)—which enabled Meta to lay the groundwork for the Metaverse.

Data center
Request a Callback

Talk to the Experts at Penguin Solutions

Connect with Penguin Solutions to design, build, deploy, and manage your NVIDIA DGX-accelerated AI clusters.

Reach out today and let's discuss your AI infrastructure requirements.

Let's Talk