Nutanix’s Kubernetes Platform: Just Another Single Pane of Glass?
BARCELONA, Spain — It was revealed during Nutanix’s annual user’s conference here, .NEXT 2024, that the company’s Nutanix Kubernetes Platform (NKP) will offer a comprehensive platform for Kubernetes management across multiclouds that does not require the usual Kubernetes know-how for individual services. AI will do much of the orchestration and management of the clusters, although a company representative did not reveal details about how this would be integrated into NKP, ahead of its expected availability this summer as part of Nutanix’s Project Beacon for Kubernetes.
@nutanix CEO @RajivRamaswami: Project Beacon now covers @kubernetesio management, infrastructure data services and platform services. #NEXTconf 2024. pic.twitter.com/55HuxVpmIG
— BC Gain (@bcamerongain) May 21, 2024
Another pane of glass that sounds too good to be true? Without mentioning any names, there are many “tombstones” out there right now, especially today offering a single do-everything platform for Kubernetes. That said, prior to six months ago, Nutanix already offered a well-tested single platform allowing organizations to adopt multicloud and infrastructure covering on-premises as well more easily. But with NKP, an entire Kubernetes management platform is on offer as well.
“The Nutanix operating model has been, for a while, to help customers run any app anywhere with a single platform. With this new product, we’re expanding that operating model to Kubernetes,” Tobi Knaup, senior director and general manager of cloud native for Nutanix, told The New Stack. “A single platform anywhere for your apps and your data. We also have a deep portfolio of data services.”
@nutanix’s Tobi Knaup (@mesosphere co-founder) on making the leap to @kubernetesio less painful? Is Nutanix Kubernetes Platform (NKP) more than just another pain of glass for Kubernetes? pic.twitter.com/66MtfDOnOS
— BC Gain (@bcamerongain) May 21, 2024
Lowering the Threshold
If what Nutanix is touting can be held to be true, it offers a way to significantly simplify and lower the threshold of Kubernetes adoption both from the user and management aspects. The encompassing Nutanix cloud stack for Kubernetes includes the major cloud providers, of course, including GCP, AWS and Azure already, but now you have Kubernetes management through these cloud providers or other platforms, including OpenShift.
This platform is a combination of Nutanix’s Kubernetes Engine and the main feature, the D2IQ product family, which Nutanix acquired in 2023. D2IQ was previously Mesosphere, which Knaup co-founded.
NKP is designed as a central platform that allows you to manage your entire fleet of Kubernetes clusters. Whether those are NKP clusters running on Nutanix infrastructure or vSphere, a platform that has been supported for years and will continue to be supported, or a public cloud-managed Kubernetes service like EKS or AKS.
NKP manages an entire fleet of clusters and embraces and extends these public cloud services with the full suite of open source add-ons that are part of NTP. Lifecycle management is used to bring up those clusters from the management plane and then build them out into a fully production-grade solution using open source components.
This is crucial because it allows customers to have a single way to operate and manage Kubernetes anywhere they run. It also means that, because it is so space-efficient, their applications remain truly portable. This is quite different from going all in on one particular cloud and using all the tools available there, which can result in being locked into that specific cloud, making your applications non-portable. We provide an alternative to that. “Portability was one of the core promises of containers and Kubernetes, and we ensure that this promise remains fulfilled,” Knaup said.
NKP will also make use of GPT-in-a-Box 2.0, which Nutanix introduced this week, for expanded Nvidia accelerated computing and LLM support. GPT-in-a-Box 2.0 will also include simplified foundational model management and integrations with Nvidia NIMs microservices and the Hugging Face LLM library. GPT-in-a-Box 2.0 will offer API endpoint creation, covering foundation model and end-user access key management.
These developments are of interest to organizations such as the University of Canberra, which is re-evaluating its cloud strategy. With Nutanix now offering the ability to run Nutanix clusters on public providers like AWS and Azure, it is considering this as part of its hybrid multi-cloud strategy, Justin Mason, associate director, vendor and operations, University of Canberra in Australia, told The New Stack. “When we talk about self-service and cloud native apps, it is about putting control back into the hands of active users,” Mason said. “This aims to provide a better user experience and more empowerment to their students and researchers.”
For instance, Nutanix offers a product called Calm. It’s similar to the blueprints available from public cloud providers, with which organizations can spin up Kubernetes clusters or other services. A researcher could go into a self-service portal, request to spin up an SQL instance, and Nutanix Calm could handle that for them, similar to a public cloud provider, Mason said.
Otherwise, with different public cloud providers, different experts are usually required for each one if something needs to be done in Azure, AWS or GCP, Mason said. “Being a small university, it is hard to have skillsets across all these diverse public cloud providers,” Mason said. “However, if they already have the Nutanix skills, then it’s easy to run. Or, if they can buy a SaaS service off the shelf, they don’t have to worry about those skill sets because it is managed for them.”
NKP Features
Specifically, Nutanix communicates in its documentation the following features for NKP:
- Kubernetes Platform for Data-Driven Apps: NKP integrates with the industry-leading Nutanix portfolio of data services, providing reliable scale-out block, file, and object storage, as well as databases-as-a-service.
- Simplified Kubernetes Management through Automation: NKP is powerful in its ability to simplify Kubernetes deployment, security, monitoring and upgrades through best-of-breed automation and AI-driven operational insights that reduce complexity and human error, and establish consistency and standardization.
- Complete Platform without Vendor Lock-In: NKP is built on CNCF-conformant Kubernetes, enabling customers to enjoy the innovation of the open-source community while avoiding the portability, compatibility, upgradeability, and security issues associated with forked Kubernetes versions and single-vendor solutions. Additionally, because NKP is part of the cloud native ecosystem, customers can easily leverage CNCF-compatible partner solutions to build a Platform Engineering stack that fits their particular needs.
- Multicluster Fleet Management: NKP eases the management of Kubernetes clusters through a centralized management plane, a user-friendly dashboard that acts as a single point of observability and control for clusters running on-premises, in public cloud, edge, and air-gapped environments. NKP extends public cloud Kubernetes services like EKS and AKS with a full suite of platform services, giving customers a single standardized way of managing their entire Kubernetes fleet.
