Data Center / Cloud

Catapulting Enterprises to the Leading Edge of AI with NVIDIA AI Enterprise 3.1

An illustration of production AI workflow applications such as video conferencing, computer vision, and robotics.

Generative AI has marked an important milestone in the AI revolution journey. We are at a fundamental breaking point where enterprises are not only getting their feet wet but jumping into the deep end.

With over 50 frameworks, pretrained models, and development tools, NVIDIA AI Enterprise, the software layer of the NVIDIA AI platform, is designed to accelerate enterprises to the leading edge of AI, while also simplifying AI to make it accessible to every enterprise.

In other exciting news is the announcement of a collaboration between AT&T Corp. and NVIDIA in which AT&T will continue to transform its operations and enhance sustainability by using AI powered by NVIDIA. This includes enhancing its data processing using the NVIDIA AI Enterprise software suite, which now includes the NVIDIA RAPIDS Accelerator for Apache Spark.

Release highlights: Accelerated data processing, AI workflows, expanded availability in the cloud

At NVIDIA GTC 2023, NVIDIA announced the latest version of NVIDIA AI Enterprise, speeding time to value across on-premises and cloud. The updates add the following features and tools and more:

  • RAPIDS Accelerator for Apache Spark
  • NVIDIA AI Workflows for personalized product recommendations and route optimization
  • Expanded availability in the cloud
  • Support for the latest NVIDIA Ada Lovelace architecture GPUs

A 5x data processing speed-up with RAPIDS Accelerator for Apache Spark

Data processing transforms datasets into meaningful and informative forms to fuel AI. But as organizations accumulate increasingly large amounts of data, the cost and time of processing grows.

NVIDIA AI Enterprise 3.1 includes RAPIDS Accelerator for Apache Spark, enabling data scientists and data engineers to accelerate Apache Spark 3.x up to 5x and save up to 75% on infrastructure costs over CPU-only platforms, without code changes.

You can get optimized performance for Spark deployments with full access to enterprise-grade support on certified platforms, including Amazon EMR, Google Cloud Dataproc, and Databricks on Azure and AWS.

New AI workflows: Next item prediction and route optimization

NVIDIA AI workflows are cloud-native, prepackaged reference examples that show how NVIDIA AI frameworks can be used to build AI solutions. Two new AI workflows are available with NVIDIA AI Enterprise 3.1:

These reference applications provide you with a path to get started quickly with AI-powered recommender and route optimization solutions and meet your business objectives.

NVIDIA AI Enterprise available on cloud marketplaces

To increase the accessibility of enterprise-supported AI software in the cloud, NVIDIA AI Enterprise is now available on the Google Cloud Marketplace and will be available soon on the AWS and Microsoft Azure Marketplaces.

You can get started right away on-demand or use your committed cloud-spend agreements to access the full-stack software suite in minutes. NVIDIA AI Enterprise includes AI frameworks and tools to build data science pipelines and AI applications with enterprise-grade support provided by NVIDIA.

It is also certified to run on Google Cloud, AWS, Azure, and Oracle Cloud for customers who bring their own licenses.

Kubernetes automation with CSP container orchestration services

NVIDIA AI Enterprise 3.1 supports Kubernetes container hosting environments, including Amazon Elastic Kubernetes Service (Amazon EKS) and Google Kubernetes Engine (GKE).

By using the integrated GPU instances, you can take advantage of a fully supported, optimized AI software stack and simple deployment to run distributed AI jobs in the cloud efficiently.

Acceleration with the latest GPU architecture, enhancement for hybrid cloud infrastructure

The following NVIDIA Ada Lovelace architecture GPUs are supported with the latest NVIDIA AI Enterprise version to supercharge computationally intensive AI and inference workloads:

  • NVIDIA L40
  • NVIDIA RTX 6000 Ada Generation

Get hands-on experience to run NVIDIA AI Enterprise on NVIDIA L4 GPUs through NVIDIA LaunchPad.

The enhancements to hybrid cloud infrastructure through this release deliver support for heterogeneous profile types for mixed workloads and MLOps certifications with ClearML. With the Update 1 release, VMware vSphere 8 is now certified and supported to enable NVIDIA NVSwitch deployment.

Get started with NVIDIA AI Enterprise

Sign up for NVIDIA LaunchPad to gain short-term access to experience the NVIDIA AI Enterprise software suite in a private accelerated computing environment that includes sets of hands-on labs. Experience NVIDIA AI workflows for product recommendations, route optimization, speech AI, cybersecurity, and more.

Watch GTC 2023 sessions

Learn the best practices and solve common challenges of AI adoption from NVIDIA, Carilion Clinic, and ControlExpert by watching the panel, A Practical Guide to AI: Moving from Pilot to Production.

Check out the AI workflow sessions where NVIDIA experts discuss session-based recommendation models and route optimization.

Discuss (0)