Securing Data With Cedar
summary
Demo plugin embeds a Cedar-based Java PDP into OpenSearch to filter Search API results per-document, mapping claims from tokens to Cedar objects and emitting concise audit logs (JTI references) to a central collector. Measured latency is ~2 ms per authorization (≈1500 µs Java binding, ≈400 µs JWT validation); plan is JNI binding + JWT caching to reach the stated TBAC with signed JWTs goal of ~50–75 µs. This is a prototype, not a supported product.
AI-written summary · about
chapters (8)
- 0:00Cedar introduction
- 1:27Why Cedar was created
- 3:58Cedar vs Rego and others
- 8:10Token-based access control
- 11:21OpenSearch plugin demo
- 15:40Performance benchmarks
- 17:23Next steps and feedback
- 28:56Q&A
related talks
52mAccess management: Customer use of Cedar policy & Verified Permissions
20mCedar: A New Language for Expressive, Fast, Safe, and Analyzable Authorization
14mOpenSearch Data Separation With JWT Authentication for Multi-tenant Applications
45mNext Generation Authorisation using Cedar
40mSecuring AI RAG Pipelines with Fine Grained Authorization
01h 00mSupercharging OAuth 2.0 Security
35mSecure JAX-RS
39mMicroservice Security via MicroProfile JWT