Interplanetary compute: The Moon, Mars, and beyond
summary
Claim: put cloud compute into orbit to cut launch-to-revenue and downlink costs. Hardware: X86 CPUs, AMD GPUs, Intel Myriad VPUs, FPGAs running Docker/TensorFlow in space since 2016. Results: cargo buses deploy satellites and shorten deployment 85% faster to revenue and cut deployment cost ~40%; one mission processes 100 GB/day to 4 GB/day and Sentinel-2 chips from 2.5 GB to 267 KB — 99.9% data reduction. Radiation voting boosts resilience ~720x (to ~5.9 years); roadmap reaches 30–50 TOPS for in-orbit retraining.
AI-written summary · about
chapters (9)
- 0:00Introduction and speakers
- 2:58Space logistics infrastructure
- 19:50Radiation-tolerant hardware
- 37:19Interplanetary communication challenges
- 40:03Data sovereignty in space
- 44:20Quantum computing impact
- 46:20Investing in space exploration
- 52:16Space debris solutions
- 55:34Closing remarks
related talks
13mSpace Clouds
55mAbove the Cloud: Building Data Centers in Space
47mBoldly Going, Running Linux in Space
06mMETASAT Demonstrator: Mixed Criticality, Accelerated AI Computing for Future Space Systems
49mHARDWARE in SPACE! The Kernel at the Edge of the Universe
16mA novel Cloud High Performance Computing infrastructure for user
17mKeynote: Beyond the Cloud: Edge Computing
29mCode the future of space