Enabling Compute and Data Sovereignty with Infrastructure-Level Data Spaces
summary
Claims an infrastructure-level dataspace that offloads computation into remote clusters using Lico-built cluster expansion, deploying an offloaded pod with a sidecar proxy that enforces traffic limits and aggregation policies. Experiments show no deployment overhead versus vanilla pod deployment, a small resource overhead during transfer, and a 37% median latency increase for dataspace transfers vs local connectivity, offset by reduced downstream bandwidth after local aggregation. Proposes integration with IDSA/Gaia-X connectors for auth and tunnel setup.
AI-written summary · about
chapters (8)
- 0:00Data sovereignty basics
- 3:15App-level data space challenges
- 4:20LiCO for flexible spaces
- 5:44Pharma-hospital data access
- 7:11Performance metrics
- 8:40Latency across solutions
- 11:10GAIA-X connector integration
- 12:38Q&A
related talks
28mRevolutionizing the Big Data Age With Compute over Data
24mHow to Drive Impact and Scale with Your Big Data/AI Applications?
07mData Gravity and Kubernetes: Managing Large-Scale Data Ingest With Minimal Latency
29mOn Data Locality in Kubernetes
22mThe Role Of A Well Implemented Data Mesh In Achieving Your ESG Responsibilities
45mEnabling Data Sovereignty in AI Workflows through Data Mesh Architecture
32mDataspaces & Gaia-X: Potentials For Trustworthy, Secure and Privacy-preserving Data-sharing
05mLightning Talk: How to Bring Data Locality to I/O-Intensive Workloads on Kubernetes