K8s meets confidential computing - the different ways of scaling sensitive workloads
summary
Edgeless Systems outlines three concrete ways to run sensitive workloads: Intel SGX for process-level enclaves (requires SDKs, device drivers, and app adaptation), AMD SEV/TDX and ARM Realms for full confidential VMs (lift‑and‑shift containers with ~single‑digit to ~9% runtime overhead), and cluster‑wide confidential Kubernetes that moves nodes — and optionally the control plane — into attested VMs. Remote attestation, CSI node‑level encryption, mutual TLS provisioning, and tools like marble run and constellation provide orchestration, key delivery, and verifiable deployment (example: Bosch privacy‑preserving AI pipeline).
AI-written summary · about
chapters (9)
- 0:00Start
- 3:33SECURE ENCLAVES
- 9:24CONFIDENTIAL CONTAINERS
- 11:54THE BIGGER PICTURE
- 12:40THE CHALLENGE
- 14:13THE SOLUTION
- 17:09KUBERNETES WITH CONFIDENTIAL VMS
- 18:00CONFIDENTIAL KUBERNETES
- 20:44OPEN SOURCE TOOLS
related talks
40mKubernetes Meets Confidential Computing - The Different Ways of Scaling Sensitive Workloads- 40mKubernetes Meets Confidential Computing - The Different Ways of Scaling Sensitive...
05mSponsored Keynote: Bringing Confidential Computing to Kubernetes
11mLightning Talk: Computing Confidentially in the Clouds
46mConfidential Cloud: Safeguarding Sensitive Data in K8s
38mCloud native Confidential: Isolating Kubernetes Pods with Confidential Computing
16mIntroduction to confidential computing
42mSponsored Session: Confidential Computing - New Capabilities for New Workloads