Introduction to confidential computing – protecting sensitive workloads
summary
Compares homomorphic encryption, secure multiparty computation and trusted execution environments with concrete details: homomorphic example encrypts 7 and 3 by multiplying with key 15 then adds remotely; HE can blow one clear bit to ~4,900 encrypted bits and needs cryptographers and verification layers; MPC splits inputs across parties, costs heavy network exchange; TEEs (Intel SGX, AMD SEV/TDX) provide attestation, secure boot and memory encryption, are the most mature today, used by Azure, AWS Nitro and limited Google offerings, but attestation and vendor lock remain concerns.
AI-written summary · about
chapters (12)
- 0:00Confidential computing intro
- 1:26Why sensitive data needs protection
- 4:55Three techniques compared
- 5:18Homomorphic encryption explained
- 12:36Secure multiparty computation in practice
- 20:29Trusted execution environments
- 24:23Secure boot and remote attestation
- 32:11Maturity of techniques compared
- 33:59AMD and Intel hardware examples
- 37:15Cloud providers' confidential offers
- 41:56Adopting confidential computing
- 43:24Q&A
related talks
40mConfidential Computing and Privacy-Enhancing Technologies - The Landscape
42mSponsored Session: Confidential Computing - New Capabilities for New Workloads
37mIt's All About Trust: Why Confidential Computing Matters
11mLightning Talk: Computing Confidentially in the Clouds
47mThe Rise of Confidential Computing, What It Is and What it Means to You- 37mIt's All About Trust: Why Confidential Computing Matters
36mEverybody is talking about Confidential Computing - the minimum Devs should know- 42mSponsored Session: Confidential Computing - New Capabilities for New Workloads