Skip to content

Details

๐Ÿ›‘ Important:
Please REGISTER HERE to join the webinar! We will host this webinar in BrightTalk.

๐Ÿ—’ Overview:
Most multi-tenant AI infrastructure relies on software overlays and trust-based controls, which can result in noisy neighbors, data leakage between tenants, and unprovable isolation guarantees. Hardware-enforced isolation prevents these issues by grounding tenant isolation across several distinct layers of the stack.

In this conversation, hear insights from Mirantis and NVIDIA leaders Shaun Oโ€™Meara and Brad Burres as they explore the growing market demand for multi-tenant AI infrastructure, and how Mirantis and NVIDIA are working together to deliver robust multi-tenancy for AI factories.

Theyโ€™ll break down the NVIDIA ecosystem primitives that make hardware-enforced multi-tenancy possible and how Mirantis k0rdent AI orchestrates them into a coherent architecture. The conversation will be moderated by Rama Darbha of NVIDIA.

Attendees will learn:

  • How Mirantis and NVIDIA are collaborating to deliver hardware virtualization (x86 and ARM) that enforces hard tenant compute boundaries at scale, meeting the performance and isolation requirements of AI factories
  • How BlueField DPUs move network policy enforcement off the host CPU, providing hardware-level tenant network segregation across the fabric
  • Where multi-tenant storage fits in the stack, operating over tenant-segregated networking to deliver per-tenant data safety at scale
  • How k0rdent AI's multi-tenant service layer leverages hardware isolation from NVIDIA Infra Controller (NICo) to deliver tenant-isolated AI services at scale
  • What tenant onboarding, mid-lifecycle management, and hardware sanitization on release look like as a repeatable operational workflow

๐Ÿ”— REGISTER HERE
๐Ÿก Location: Online (youโ€™ll get the link once you register)
โฐ Date/Time: Tuesday, Aug 25 @ 9 am PT / 12 pm ET / 6 pm CEST
๐Ÿ‘‰๐Ÿป Speaker: Shaun Oโ€™Meara, CTO at Mirantis & Brad Burres, Senior Distinguished Engineer at NVIDIA
๐Ÿ‘‰๐Ÿป Moderator: Rama Darbha, Sr. Director of Solutions Architecture at NVIDIA

Related topics

AI/ML
Cloud Computing
Cloud Infrastructure
Cloud Services
Nvidia

You may also like