Skip to main content
Converged Infrastructure

FlexPod Datacenter with generative AI inferencing

FlexPod Datacenter is the intelligent platform for GPU-intensive workloads. Cisco, NetApp, and NVIDIA have collaborated to create a design and deployment guide that seamlessly integrates Cisco UCS Servers, Cisco Nexus switches, and NetApp storage with NVIDIA GPUs.

Request Pricing View Datasheet
FlexPod Datacenter with generative AI inferencing

Artificial intelligence workloads generate massive amounts of structured and unstructured data. Cisco, NetApp, and NVIDIA have collaborated to create a design and deployment guide that seamlessly integrates Cisco UCS Servers, Cisco Nexus switches, and NetApp® storage with NVIDIA GPUs. The FlexPod® Cisco Validated Design (CVD) provides a comprehensive and streamlined approach for organizations to configure and customize their environments to provide robust performance for virtually any AI workload.

FlexPod Datacenter with generative AI inferencing overview

Features and Benefits

The capabilities that set FlexPod Datacenter with generative AI inferencing apart.

The intelligent platform for GPU-intensive workloads

Powered by fourth-generation Xeon Scalable processors

Cisco x210c M7 compute nodes provide up to 120 cores and up to 8TB of RAM per server, which work in tandem with Cisco UCS X440p PCI Express nodes.

NVIDIA A100-80 GPUs

Each PCIe node houses NVIDIA A100-80 GPUs, connecting to compute nodes over Cisco UCS X-Fabric.

NetApp AFF A800 all-flash storage

The CVD validates the NetApp AFF A800 all-flash storage system, with NetApp Astra Trident™ layered on top of NVMe-TCP with 100Gbe providing high-performance persistent storage for containerized workloads.

Cisco Nexus 9000 series switches and Cisco UCS 6500 series fabric interconnects

Joining the compute and storage are the latest Cisco Nexus 9000 series switches and Cisco UCS 6500 series fabric interconnects. This combination provides the high performance that generative AI inferencing software and models require.

Programmable infrastructure simplifies deployments

Design and deployment in one document

The Cisco Validated Design covers both the design and deployment in one document. By referencing the Infrastructure as Code (IaC) CVD, Ansible is used to simplify the deployment of Red Hat OCP on top of vSphere 8.

3-node management cluster with 5 nodes for AI workloads

This results in a 3-node management cluster with 5 nodes to handle containerized AI workloads. Each large language and image generation model runs as a container, creating a persistent volume claim through Astra Trident, which enables persistent, high-performing storage direct to the model of choice.

Cisco Intersight enables sustainable visibility

Cisco Intersight enables sustainable visibility

Benchmarks for leading models

The FlexPod Generative AI CVD covers benchmarks for some of the industry's leading large language and image generation models, demonstrating low latency and high-performance inferencing.

Power & Energy Metrics Dashboard

Monitoring this performance is easy with Cisco Intersight, which offers a Power & Energy Metrics Dashboard. This dashboard offers a high-level overview of the top five power-consuming blade and rack servers, as well as individual host power usage refreshed by the minute to monitor power consumption the moment your AI workloads kick off.
Expert Guidance

Thrive with expert-led storage guidance

Get tailored advice on how FlexPod Datacenter with generative AI inferencing fits your environment — from sizing and deployment to long-term optimization.

Talk to a specialist
Thrive with expert-led storage guidance

Technical Specifications

Exhaustive hardware and software metrics extracted directly from official documentation.

  • Compute Nodes
    Cisco x210c M7
  • Processors
    Fourth-generation Xeon Scalable processors
  • Cores per Server
    Up to 120 cores
  • RAM per Server
    Up to 8TB
  • PCIe Nodes
    Cisco UCS X440p PCI Express nodes
  • GPUs
    NVIDIA A100-80
  • GPU Connectivity
    Cisco UCS X-Fabric

  • Storage System
    NetApp AFF A800 all-flash
  • Storage Orchestration
    NetApp Astra Trident™
  • Storage Protocol
    NVMe-TCP with 100Gbe

  • Switches
    Cisco Nexus 9000 series
  • Fabric Interconnects
    Cisco UCS 6500 series

  • Deployment Automation
    Ansible (Infrastructure as Code)
  • Container Platform
    Red Hat OCP
  • Hypervisor
    vSphere 8
  • Cluster Configuration
    3-node management cluster with 5 nodes for containerized AI workloads
  • Monitoring
    Cisco Intersight Power & Energy Metrics Dashboard

Ready to get started?

Get your data flowing from edge to core to cloud.

Talk to a specialist

Request a custom quote

Build a configuration with a Converged Infrastructure specialist.

Request a quote

Download the datasheet

Full specs, performance metrics, and deployment notes.

Get the datasheet

Learn more

Explore resources

Datasheets, whitepapers, case studies, and technical documentation.

Explore resources

View solutions

Tailored storage and data management solutions for your workloads.

View solutions

Most secure storage on the planet FIPS 140-3 · NSA CSfC · DoDIN APL
Validated for top-secret data Only enterprise storage to hold this certification
Authorized NetApp Partner SANDataWorks · a division of BlueAlly
Free Security Assessment