When the Universe Calls
summary
History shows early probes ran on 512 bytes → 128 22‑bit words but modern missions dump terabytes daily: SDO hit ~5 TB/day, Gaia sends ~70 GB/day, Event Horizon Telescope recorded ~64 GB/s. Projects built: Galaxy Zoo crowdsourced millions of labels, enabling CNNs to classify SDSS and DES after image rescaling; Moon Mappers (launched 2012, PHP→Python) used volunteer crater maps to train TensorFlow models; OSIRIS‑REx maps Bennu (~500 m) with human-marked boulders because current ML still fails on varying sun angles, soils, and scales. MariaDB on AWS RDS and GPUs power the pipeline.
AI-written summary · about
chapters (12)
- 0:00Start
- 3:51Tornados of Plasma on the Sun
- 4:32Gaia Spacecraft
- 13:32The Dark Energy Survey
- 14:52The Moon
- 16:08Apollo 14
- 19:02Moon Mappers
- 23:34Angle Matters
- 27:02Osiris Rex
- 43:40Main Goal
- 45:01Evolution of Lakes on Glaciers
- 46:29Gamification Strategies
related talks
05mAlex Kudriashova Interview
36mThe (Hello) World, The Universe and Uranus
20mMining Large Data Sets of Biomedical Omics Da...
17mIt's All in the Cloud
11mProcessing TBs of Astronomy Data at Scale with the AC3 Horizon Research Project
28mAdvancing the future of space in the cloud
46mChasing the Dark Universe with Euclid and Python: Unveiling the Secrets of the Cosmos
36mRapid Processing of NASA Satellite Data Stored with Ceph