As the world transitions from general-purpose to accelerated computing, finding a path to building data center infrastructure at scale is becoming more important than ever. Enterprises must navigate uncharted waters when designing and deploying infrastructure to support these new AI workloads.
Constant developments in model capabilities and software frameworks, along with the novelty of these workloads, mean best practices and standardized approaches are still in their infancy. This state of flux can make it difficult for enterprises to establish long-term strategies and invest in infrastructure with confidence.
To address these challenges, NVIDIA is unveiling Enterprise Reference Architectures (Enterprise RAs). These comprehensive guidelines help NVIDIA systems partners and joint customers build their own AI factories — high-performance, scalable and secure data centers for manufacturing intelligence.
Building AI Factories to Unlock Enterprise Growth
NVIDIA Enterprise RAs help organizations avoid pitfalls when designing AI factories by providing full-stack hardware and software recommendations, and detailed guidance on optimal server, cluster and network configurations for modern AI workloads.
Enterprise RAs can reduce the time and cost of deploying AI infrastructure solutions by providing a streamlined approach for building flexible and cost-effective accelerated infrastructure, while ensuring compatibility and interoperability.
Each Enterprise RA includes recommendations for:
- Accelerated infrastructure based on an optimized NVIDIA-Certified server configuration, featuring the latest NVIDIA GPUs, CPUs and networking technologies, that’s been tested and validated to deliver performance at scale.
- AI-optimized networking with the NVIDIA Spectrum-X AI Ethernet platform and NVIDIA BlueField-3 DPUs to deliver peak network performance, and guidance on optimal network configurations at multiple design points to address varying workload and scale requirements.
- The NVIDIA AI Enterprise software platform for production AI, which includes NVIDIA NeMo and NVIDIA NIM microservices for easily building and deploying AI applications, and NVIDIA Base Command Manager Essentials for infrastructure provisioning, workload management and resource monitoring.
Businesses that deploy AI workloads on partner solutions based upon Enterprise RAs, which are informed by NVIDIA’s years of expertise in designing and building large-scale computing systems, will benefit from:
- Accelerated time to market: By using NVIDIA’s structured approach and recommended designs, enterprises can deploy AI solutions faster, reducing the time to achieve business value.
- Performance: Build upon tested and validated technologies with the confidence that AI workloads will run at peak performance.
- Scalability and manageability: Develop AI infrastructure while incorporating design best practices that enable flexibility and scale and help ensure optimal network performance.
- Security: Run workloads securely on AI infrastructure that’s engineered with zero trust in mind, supports confidential computing and is optimized for the latest cybersecurity AI innovations.
- Reduced complexity: Accelerate deployment timelines, while avoiding design and planning pitfalls, through optimal server, cluster and network configurations for AI workloads.
Availability
Solutions based upon NVIDIA Enterprise RAs are available from NVIDIA’s global partners, including Dell Technologies, Hewlett Packard Enterprise, Lenovo and Supermicro.
Learn more about NVIDIA-Certified Systems and NVIDIA Enterprise Reference Architectures.