SKA Observatory: Reaching to the Stars and Beyond
summary
Global radio observatory spans three sites (UK HQ, South Africa, Australia) and targets first community science data in 2026 while producing on the order of hundreds of petabytes per year, forcing on‑the‑fly processing. Data architecture requires multi‑primary replication, low local latency, independent site operation and acceptably stale reads — roughly follower reads with ~15s staleness — for outage resilience. Evaluation favors YugabyteDB for geo‑partitioning, follower reads, Kubernetes operator and Postgres compatibility to implement on‑prem multi‑region distribution.
AI-written summary · about
chapters (16)
- 0:00SKA Observatory intro
- 2:15Global astronomy landscape
- 3:38Global partnership and scale
- 5:02Telescope locations and infrastructure
- 7:15Science data processing challenges
- 8:21Construction timeline and milestones
- 10:07Current deployment and future plans
- 11:35Construction activities and site support
- 12:21Software architecture and control systems
- 13:46Database requirements and solutions
- 15:52Data replication and distribution challenges
- 17:16PostgreSQL multi-primary replication
- 19:03Geographically distributed data needs
- 20:30YugabyteDB for SKA evaluation
- 22:42YugabyteDB features and benefits
- 24:10Q&A
related talks
06mLightning Talk: SKAO: Next-Generation Radio Astronomy-Driven Big Data
18mBuilding a Scalable, Resilient System for Real-Time Fraud Detection
37mLarge Synoptic Survey Telescope: From Cloud Native to Dark Energy
19mBuild Global-Scale Apps Using Globally Distributed Autonomous Databases
24mScaling to Billions of Internet Observations per Day
39mEdge Compute Infrastructure for Experience Business
45mFrom telescope to data centre: adventures in astronomical data pipelines
30mFrom monolith to petabyte scale: Built a cloud-native, AI-ready data layer