Towards the medicine of the future with confidential computing
summary
DigiMed Bayern builds an academic confidential cloud at LRZ to let 14 institutions and >100 researchers share multi‑omic heart data with €25M funding; the stack uses OpenStack, core‑byte file system and AMD SEV‑based enclaves, end‑to‑end encrypted multi‑tenancy and a Black Box thread model. A full NGS pipeline and an on‑device ADSFit app are migrating to the cloud without code changes. It works technically but adoption, legal frameworks and staffing remain the main hurdles.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:46Bavarian Cloud for health research
- 2:51Bavarian Ministry of Health
- 4:39Data set
- 6:39The Cloud
- 8:41Examples
- 10:51Lessons learned
related talks
24mAI, Health & Confidentiality
30mSharing without Sharing: the Case for Data Clean Rooms Powered by Confidential Computing
13mProvably Secure AI for Health Data – Powered by Confidential Computing
44mAdvancements in Confidential Computing
49mCloud Security Made for the EU: Securing Data & Applications
37mCloud, security, and science converge in a data-driven world (PRT324)
20mThe Transformational Power of Confidential Computing
13mVM Security with Confidential Computing