Trusted Compute integration to Smart Intersection

[SupriyaPamulpati]

Description

Trusted Compute (TC) Overview

Trusted Compute (TC) is a workload protection solution that leverages Intel Platform hardware capabilities. For detailed information, please refer to: https://docs.openedgeplatform.intel.com/2026.0/OEP-articles/application-security/trusted_compute_introduction.html

Recommended Use Case

TC is the recommended solution for protecting high-sensitivity assets such as AI models and AI pipelines. We propose deploying the DLSPS and VLM Serving microservices within the TC environment to provide runtime isolation and execution protection. This ensures that AI models and pipelines loaded into memory by these microservices remain isolated from other workloads and microservices running on the same device.

PR Changes Summary

This PR introduces modifications to the deployment of DLSPS deployment:

1. Added kata-qemu value to the runtimeClass annotation. This annotation instructs containerd to use kata-qemu instead of runc, enabling kata-qemu to create VM sandboxes for DLSPS and VLM Serving containers.
2. Added explicit CPU and memory resource allocation requirements for executing DLSPS in VM sandbox environments. These resource definitions enable the runtime to create VM sandboxes with the appropriate resources needed for the application.

Fixes # (issue)
NA

Any Newly Introduced Dependencies

NA

How Has This Been Tested?

On the ArrowLake & PantherLake system using CPU compute. TC support Application usage of GPU device however in this version we have not validated. We will add the GPU support in future.

Checklist:

• I agree to use the APACHE-2.0 license for my code changes.
• I have not introduced any 3rd party components incompatible with APACHE-2.0.
• I have not included any company confidential information, trade secret, password or security token.
• I have performed a self-review of my code.

[shahmitu]

remove the document. This duplicate of Tutorial.