This optimizes the use of the GPU hardware and it can serve more than one user, reducing costs. A basic level of familiarity with the core concepts in Kubernetes and in GPU Acceleration will be useful to the reader of this article. We first look more closely at pods in Kubernetes and how they relate to a GPU. A pod is the unit of deployment, at the lowest level, in Kubernetes. A pod can have one or more containers within it. The lifetime of the containers within a pod tend to be about the same, although one container may start before the others, as the "init" container. You can deploy higher-level objects like Kubernetes services and deployments that have many pods in them. We focus on pods and their use of GPUs in this article. Given access rights to a Tanzu Kubernetes cluster (TKC) running on the VMware vSphere with Tanzu environment (i.e. a set of host servers running the ESXi hypervisor, managed by VMware vCenter), a user can issue the command:

VMware vSphere with Tanzu provides users with the ability to easily construct a Kubernetes cluster on demand for model development/test or deployment work in machine learning applications. These on-demand clusters are called Tanzu Kubernetes clusters (TKC) and their participating nodes, just like VMs, can be sized as required using a YAML specification. In a TKC running on vSphere with Tanzu, each Kubernetes node is implemented as a virtual machine. Kubernetes pods are scheduled onto these nodes or VMs by the Kubernetes scheduler running in the Control Plane VMs in that cluster. To accelerate machine learning training or inference code, one or more of these pods require a GPU or virtual GPU (vGPU) to be associated with them.

Enterprises can begin to run trials of their AI projects using VMware vSphere with Tanzu together with Nvidia AI Enterprise software suite, as part of moves by both companies to further simplify AI development and application management. By extending testing to vSphere with Tanzu, Nvidia boasts it will enable developers to run AI workloads on Kubernetes containers within their existing VMware environments. The software suite will run on mainstream Nvidia-certified systems, the company said, noting it would provide a complete software and hardware stack suitable for AI development. "Nvidia has gone and invested in building all of the next-generation cloud application-level components, where you can now take the NGC libraries, which are container-based, and run those in a Kubernetes orchestrated VMware environment, so you're getting the ability now to go and bridge the world of developers and infrastructure," VMware cloud infrastructure business group marketing VP Lee Caswell told media. The move comes off the back of VMware announcing Nvidia AI Enterprise in March.

Hundreds of thousands of companies worldwide will now have the ability to run AI on VMware vSphere and industry-standard servers thanks to NVIDIA software. A comprehensive software set of AI tools and frameworks is now available from NVIDIA, enabling VMware vSphere users to virtualize AI workloads on NVIDIA-Certified SystemsTM. During the epidemic, companies are adopting AI more and more as they realize the benefits of automation and big data analytics. AI is vital to their digital transformation initiatives. According to a separate McKinsey survey, 30 percent of firms are running AI pilots, and nearly half have integrated at least one AI capability into their typical business operations.

NVIDIA today announced the availability of NVIDIA AI Enterprise, a comprehensive software suite of AI tools and frameworks that enables the hundreds of thousands of companies running VMware vSphere to virtualize AI workloads on NVIDIA-Certified Systems . Leading manufacturers Atos, Dell Technologies, GIGABYTE, Hewlett Packard Enterprise, Inspur, Lenovo and Supermicro are offering NVIDIA-Certified Systems optimized for AI workloads on VMware vSphere with NVIDIA AI Enterprise. Separately, Dell Technologies today announced Dell EMC VxRail as the first hyperconverged platform to be qualified as an NVIDIA-Certified System for NVIDIA AI Enterprise. To help teams of data scientists run their AI workloads most efficiently, Domino Data Lab today announced it is validating its Domino Enterprise MLOps Platform with NVIDIA AI Enterprise, which runs on mainstream NVIDIA-Certified Systems. "The first wave of AI has been powered by specialized infrastructure that focused adoption on industry pioneers," said Manuvir Das, head of Enterprise Computing at NVIDIA.

This article directs you to a recent webinar that VMware produced on the topic of executing distributed machine learning with TensorFlow and Horovod running on a set of VMs on multiple vSphere host servers. Many machine learning problems are tackled using a single host server today (with a collection of VMs on that host). However, when your ML model or data grows too large for one host to handle, or your GPU power happens to be dispersed across several physical host servers/VMs, then distribution is the mechanism used to tackle that scenario. The VMware webinar introduces the concepts of machine learning in general first. It then gives a short description of Horovod for distributed training and explains the importance of low latency networking between the nodes in the distributed model, based here on Mellanox RDMA over Converged Ethernet (RoCE) technology.

NVIDIA's virtual GPU (vGPU) technology, which has already transformed virtual client computing, now supports server virtualization for AI, deep learning and data science. Previously limited to CPU-only, AI workloads can now be easily deployed on virtualized environments like VMware vSphere with new vComputeServer software and NVIDIA NGC. Through our partnership with VMware, this architecture will help organizations to seamlessly migrate AI workloads on GPUs between customer data centers and VMware Cloud on AWS. IT administrators can use hypervisor virtualization tools like VMware vSphere, including vCenter and vMotion, to manage all their data center applications, including AI applications running on NVIDIA GPUs. These GPU servers are often isolated, with the need to be managed separately.

Are you being asked to provide GPUs to your application developers and data scientists for machine learning or high performance computing? Are users asking for more than one GPU to be usable for their application? Are you interested in cost-effective ways to share GPUs across the entire data science team? If any of these types of questions apply to you, then this new E-Book from VMware on the key decisions to take about GPU use on vSphere will be a great read for you. GPUs provide the computing power needed to run machine learning programs efficiently, reliably and quickly.

VMworld 2019 US and Europe events feature many opportunities to learn about the latest in VMware vSphere server virtualization technology and operations. This page is a quick reference to the VMworld 2019 sessions and other events where customers are able to engage with VMware experts on a range of topics, as well as network with industry peers. Links to the EU sessions will be added to the coming days. You can also still access the presentations, recordings, and session information from last year here – VMworld 2018 Archive. How PowerCLI Makes vSphere Configuration Management Easy Level 300 – [US: CODE2214U] Configuration management is a key DevOps principle. PowerShell and PowerShell DSC are easy ways to make use of config management in your environment. However, there's one area that's been missing that ability: VMware. PowerCLI has introduced the key to close that gap, and it's open-sourced! The Art of Code That Writes Code Level 300 – [US: CODE2216U] REST APIs are everywhere these days. A majority of those are backed by what's known as OpenAPI (swagger) specifications. Using the vast ecosystem of OpenAPI tooling, we can generate documentation, SDKs, and even PowerShell modules.

This brief article introduces a short 4.5 minute video that explains the reasons why VMware vSphere is a great platform for data scientists/engineers to use as their base operating platform. The video then demonstrates an example of this, showing a data scientist conducting a modeling experiment with an input set of data, while using the Driverless AI tool from H2O.ai to do the data analysis and model training, all in VMs. The key idea here is that the world of machine learning/data science is rapidly changing, with new, powerful tools, platforms and versions appearing and upgrading at a very fast pace. The tool vendors are racing to innovate here and producing new workbenches for the both the expert and the novice in the field. Data scientists and data engineers (who organize and cleanse the data first) want to be able to try out these new tools and updated versions of the tools while keeping a stable environment for their existing production deployments.