Blog

HPC

Latest Version of NVIDIA AI Enterprise Accelerates AI Development and Deployment

January 19, 2022
4 min read
EXX-Blog-NVIDIA-AI-ENTERPRISE-AI-DEV-DEPLOYMENT.jpg

Artificial intelligence (AI) is becoming more and more ubiquitous—from smart personal assistants to process automation to predictive maintenance, and so much more. Many organizations know they need to implement it but face challenges carrying the project from proof of concept to deployment. 

To help enterprises get over the hurdle of pulling together an AI solution from disparate products and integrating them with existing infrastructure, NVIDIA® and VMWare® have partnered to offer NVIDIA AI Enterprise software suite that unlocks the power of AI by delivering an end-to-end enterprise platform optimized for AI workloads.

NVIDIA AI Enterprise 1.1

NVIDIA AI Enterprise is an end-to-end, cloud native suite of AI and data science applications and frameworks optimized and exclusively certified by NVIDIA to run on VMware vSphere® with NVIDIA-Certified Systems. It includes key enabling technologies and software from NVIDIA for rapid deployment, management, and scaling of AI workloads in the modern hybrid cloud. 

The latest version of the software suite brings new updates that include production support for containerized AI with the NVIDIA software and VMWare vSphere® with VMWare Tanzu™. Previously only available on a trial basis, VMWare vSphere with Tanzu™ is the industry’s leading virtualization platform that transforms bare-metal servers into centrally managed AI and machine learning infrastructure pools that enables developers to run AI workloads on Kubernetes containers within their VMWare environments.

Image Source: NVIDIA

NVIDIA AI Enterprise 1.1 software suites makes it possible for enterprises to finally run advanced AI workloads on mainstream accelerated servers, which in turn helps organizations of all types add new AI-powered services that would have previously required specialized infrastructure or operations support. 

To further simplify enterprise AI development, NVIDIA AI Enterprise also introduces support for NVIDIA RAPIDS as an integrated machine learning and deep learning container that includes TensorFlow and PyTorch libraries, as well as leading data science frameworks like Apache Spark and cuPY. RAPIDS also leverages NVIDIA CUDA® to accelerate workflows by running the entire data science training pipeline on GPUs, thereby reducing the training time and frequency of model deployment from days to minutes. And since RAPIDS can be run anywhere—cloud or on-prem—scaling from a workstation to multi-GPU servers to multi-node clusters is a breeze.


Interested in an AI workstation?

Accelerate your research with an Exxact solution starting around $5,500


NVIDIA-Certified Systems™

Exxact offers a wide variety of NVIDIA-Certified Systems that run NVIDIA AI Enterprise for a broad range of uses, including accelerated versions of the most popular models in data centers today. These NVIDIA-Certified Systems are tested with the most powerful enterprise NVIDIA Ampere architecture-based GPUs and NVIDIA Mellanox® networking that includes NVIDIA ConnectX® SmartNICs and NVIDIA BlueField® data processing unit (DPU). They are evaluated by NVIDIA engineers to ensure enterprises have the performance, functionality, scalability, and reliability they need to deliver cutting-edge products and services, while increasing operational efficiencies. 

NVIDIA-Certified Systems have been proven to deliver predictable performance and enable enterprises to quickly deploy optimized platforms for AI, Data Analytics, HPC, and other accelerated workloads.


Have any questions?
Contact Exxact Today


Free Resources

Browse our whitepapers, e-books, case studies, and reference architecture.

Explore
EXX-Blog-NVIDIA-AI-ENTERPRISE-AI-DEV-DEPLOYMENT.jpg
HPC

Latest Version of NVIDIA AI Enterprise Accelerates AI Development and Deployment

January 19, 2022 4 min read

Artificial intelligence (AI) is becoming more and more ubiquitous—from smart personal assistants to process automation to predictive maintenance, and so much more. Many organizations know they need to implement it but face challenges carrying the project from proof of concept to deployment. 

To help enterprises get over the hurdle of pulling together an AI solution from disparate products and integrating them with existing infrastructure, NVIDIA® and VMWare® have partnered to offer NVIDIA AI Enterprise software suite that unlocks the power of AI by delivering an end-to-end enterprise platform optimized for AI workloads.

NVIDIA AI Enterprise 1.1

NVIDIA AI Enterprise is an end-to-end, cloud native suite of AI and data science applications and frameworks optimized and exclusively certified by NVIDIA to run on VMware vSphere® with NVIDIA-Certified Systems. It includes key enabling technologies and software from NVIDIA for rapid deployment, management, and scaling of AI workloads in the modern hybrid cloud. 

The latest version of the software suite brings new updates that include production support for containerized AI with the NVIDIA software and VMWare vSphere® with VMWare Tanzu™. Previously only available on a trial basis, VMWare vSphere with Tanzu™ is the industry’s leading virtualization platform that transforms bare-metal servers into centrally managed AI and machine learning infrastructure pools that enables developers to run AI workloads on Kubernetes containers within their VMWare environments.

Image Source: NVIDIA

NVIDIA AI Enterprise 1.1 software suites makes it possible for enterprises to finally run advanced AI workloads on mainstream accelerated servers, which in turn helps organizations of all types add new AI-powered services that would have previously required specialized infrastructure or operations support. 

To further simplify enterprise AI development, NVIDIA AI Enterprise also introduces support for NVIDIA RAPIDS as an integrated machine learning and deep learning container that includes TensorFlow and PyTorch libraries, as well as leading data science frameworks like Apache Spark and cuPY. RAPIDS also leverages NVIDIA CUDA® to accelerate workflows by running the entire data science training pipeline on GPUs, thereby reducing the training time and frequency of model deployment from days to minutes. And since RAPIDS can be run anywhere—cloud or on-prem—scaling from a workstation to multi-GPU servers to multi-node clusters is a breeze.


Interested in an AI workstation?

Accelerate your research with an Exxact solution starting around $5,500


NVIDIA-Certified Systems™

Exxact offers a wide variety of NVIDIA-Certified Systems that run NVIDIA AI Enterprise for a broad range of uses, including accelerated versions of the most popular models in data centers today. These NVIDIA-Certified Systems are tested with the most powerful enterprise NVIDIA Ampere architecture-based GPUs and NVIDIA Mellanox® networking that includes NVIDIA ConnectX® SmartNICs and NVIDIA BlueField® data processing unit (DPU). They are evaluated by NVIDIA engineers to ensure enterprises have the performance, functionality, scalability, and reliability they need to deliver cutting-edge products and services, while increasing operational efficiencies. 

NVIDIA-Certified Systems have been proven to deliver predictable performance and enable enterprises to quickly deploy optimized platforms for AI, Data Analytics, HPC, and other accelerated workloads.


Have any questions?
Contact Exxact Today


Free Resources

Browse our whitepapers, e-books, case studies, and reference architecture.

Explore