VMware yesterday took the wraps off Project Monterey, which aims to do for hardware what Project Pacific — what later became vSphere with Tanzu — did for Kubernetes.

While just announced this week, Paul Turner, VP of product management for VMware's Cloud Platform Business Unit, said the project has actually been in the works for nearly three years. It was motivated by the adoption of disaggregated, cloud-native architectures, he explained.

This shift to Kubernetes and a development operations-heavy approach has made applications more dynamic and agile, but it "means that the infrastructure needs are dynamic too," Tuner said.

Where Project Pacific set out to integrate Kubernetes into the core of vSphere, Project Monterey aims to provide a consistent management platform for distributed compute across multiple hardware accelerators using smartNICs like those supplied by Intel and Nvidia.

"Development paradigms have already moved and changed, and they are moving to an agile DevOps style development environment, and Kubernetes is the lead for that," he said. "Today most of that is actually deployed on VMs... because that's the container that allows us to virtualize the infrastructure."

"I expect that to continue to be the case, but we want to make the VM and the resource management that VM consumes more dynamic," he said.

VMware targeted smartNICs because it's an inline device that can sit in between the server and the network and stitch together disaggregated pools of resources within a data center, Turner explained.

However, according to Turner, smartNICs are a bit of a misnomer.

"It's got the word smart in front of it, but actually it's not terribly smart," he said. "What's smart about it is I can put processing on it... and I can do manipulation of and management of the kind of traffic."

SmartNICs really start to become interesting when a management player is put in place to do resource management. "The pooling, the aggregation, the firewalling, that's where we see the big opportunity for what we're doing with Monterey."

Turner argues that, right now, smartNICs are being used almost exclusively for network monitoring and performance offload.

"We think that's really myopic on the capability there," he said. "By putting ESXi managed through vCenter, we can start building a distributed virtual machine. To do that, we need to have a control software that actually can manage aggregating and pooling of resources."

What this translates to are benefits for customers across the board, claims Turner who sees large enterprises as the biggest beneficiary of the technology.

The advantages for enterprises are going to be "improve your efficiencies on your current running environment, because you'll be able to have more VMs running on the same infrastructure you already have because you're offloading resources," he said. "The second big benefit is going to be unified management of your bare-metal environment and your virtualized environment."

For cloud providers, Turner sees the opportunity to leverage management of accelerators to enable compossible infrastructure in a way that's dynamic.

Meanwhile, Turner says Project Monterey poses a different opportunity for network operator customers.

"The telco interest in Project Monterey is that I now have an inbound optimization service that actually can truly accelerate my network IO, my network performance by network monitoring," he said. "To them, it's all about speeds and feeds."