Hybrid multicloud computing software provider Nutanix is using its newly acquired Kubernetes-expertise — thanks to last year’s purchase of D2IQ — to help customers get a jump-start on AI operations.
The company has extended its AI infrastructure platform with a new cloud native offering, Nutanix Enterprise AI, which can be deployed on any Kubernetes platform, at the edge, in core data centers and on public cloud services like AWS EKS, Azure AKS, and Google GKE.
The company plans to launch the new offering Tuesday at KubeCon+Cloud NativeCon North America 2024, held this week in Salt Lake City.
What Is Nutanix Enterprise AI?
Although most well-known as a data storage platform provider, Nutanix gained a considerable set of technologies and a virtualization platform with the purchase of D2IQ, formerly known as Mesosphere, which offered an alternative to Docker and Kubernetes, though ultimately adapted both of those technologies for its own data center-scaled platform.
“I think the combination of the two companies, of the two product lines, really addresses a strong need in the market,” said Tobi Knapp, Nutanix general manager for cloud native operations, and former CEO/co-founder of Mesophere, in an interview with TNS.
And that combined expertise is taking on one of the largest projects in IT operations, setting up AI operations for the enterprise.
Nutanix Enterprise AI is designed to deploy, run and scale inference endpoints for large language models (LLMs), allowing organizations to deploy generative AI applications in a fraction of the time it would have taken to build a deployment workflow from scratch.
Generative AI is an inherently hybrid workload, Nutanix contends. An application can be run on the cloud, but use models built from data stored on premises. Inferencing needed to determine how to best serve the customer may happen at the edge.
Such a workflow can be a challenge to set up for an organization given all the organizational (not to mention networking) boundaries it can cross.
This distributed hybrid GenAI workflow can present challenges for organizations concerned about complexity, data privacy, security, and cost.
“So the challenge we’ve identified that exists in most organizations is really the running models for the data in a safe and secure way,” Knapp said. So, the company’s goal, “is to really help organizations run models for inference wherever their data sits, and make that very easy. ”
How Can Nutanix Enterprise AI Be Used?
The software can be used in myriad ways. It can scan customer feedback and documents to find areas of dissatisfaction. It can detect fraud and create appropriate policies on the fly. It can aid in content creation through “copilots” and document processing. It can be used to fine-tune models on domain-specific data.
Nutanix Enterprise AI promises to provide a consistent multicloud operating model and works with many AI-supporting applications, such as the NVIDIA NIM optimized inference microservices and the open source Hugging Face foundational mode.
“By providing a consistent experience from the enterprise to public cloud, Nutanix Enterprise AI aims to provide a user-friendly infrastructure platform to support organizations at every step of their AI journey, from public cloud to the edge,” said Dave Pearson, Infrastructure Research VP at IDC, in a statement.
Nutanix Enterprise AI is a component of a larger offering, Nutanix GPT-in-a-Box 2.0, which also includes Nutanix Cloud Infrastructure, Nutanix Kubernetes Platform and Nutanix Unified Storage.
The idea with the platform was that it is designed to reduced a number of friction points that enterprises have encountered in the attempts to spin up genAI apps.
It addresses AI skill shortages, mitigates data privacy and security concerns and applies enterprise-wide controls to generative AI applications.
Those at KubeCon can learn more, by stopping by the Nutanix booth, A4.
The post Nutanix Extends AI Platform to Public Cloud Kubernetes appeared first on The New Stack.
Nutanix Enterprise AI now provides a platform to run GenAI apps on public clouds, in addition to running them on-premises and at the edge.