Randy Meyer, HPE & Paul Shellard, University of Cambridge | HPE Discover 2017 Madrid
summary
Cosmology group runs large-scale pattern recognition on a shared-memory HPE Superdome Flex that holds up to 40 terabytes of high-speed memory so pipelines developed on a laptop can scale without rewriting for distributed IO. They process observational data spanning 14 billion years and add gravitational-wave events discovered in 2015, using in‑memory analysis and in‑situ visualization to find rare signatures, reduce post‑processing, and iterate models much faster.
AI-written summary · about
chapters (15)
- 0:00Intro
- 0:50Computing meets the Cosmos
- 2:48What have you learned
- 3:47Sources of data
- 4:45Gravitational waves
- 6:05Multi messenger astronomy
- 7:07Observational technology
- 9:46HPE Superdome Flex
- 13:20Memory architectures
- 14:38Challenges of computing
- 15:20Simplification
- 16:20Computational power
- 17:05Cosmic microwave background
- 18:03Cosmic radiation background
- 19:11Resources
