Running Geospatial Workloads on AWS
summary
Cloud workflows flip the model so compute moves to storage: S3 hosts massive open datasets and services can query S3 with Athena or run Spark/EC2 jobs or serverless Lambdas without copying files. Examples include a 10 trillion–point LiDAR cloud, the NREL wind dataset of about 500 petabytes of HDF5 and the BlueDot Sentinel pipeline that processes a month of tiles for ~€6. Tools shown include h5py/PI-WWTK and WinVis for parallel HDF5 access and visualization.
AI-written summary · about
chapters (14)
- 0:00Intro
- 1:54Key phrase
- 2:51Open data
- 5:12Public data
- 5:55How to find data
- 6:33How to share data
- 8:52Technical considerations
- 11:05BlueDot Observatory
- 12:06LiDAR Point Cloud
- 12:36NREL
- 13:48Wind Toolkit
- 14:57H5 PYD
- 16:41Use Cases
- 17:17Key Differences
related talks
48mAccelerate science, research, and innovation with AWS
28mAdvancing the future of space in the cloud
27mVisualizing Cloud Systems
20mHow the cloud changed the way I do analytics and research
12mWhat our customers are doing with cloud outside the datacenter
28mTurn Data Into Insights
05mAWS On Air ft. Amazon EMR Serverless Quick Tip 2
17mIt's All in the Cloud