What should this benchmark look like? - Githubissues

stac-utils / stac-api-benchmark

STAC API Benchmark

Other

8 stars 2 forks source link

What should this benchmark look like? #1

Open philvarner opened 2 years ago

philvarner commented 2 years ago

A few ideas:

Datasets:

Landsat 8
MODIS MCD43A4
Sentinel 2 L2A
Sentinel 1 GRD (non-gridded)
randomly-generated non-gridded of varying sizes

Queries:

smaller than grid size
larger than grid size
all GeoJSON geometries (covers bbox, which is just a 5pt polygon rectangular aligned to the 4326 grid)

Shapes:

each country boundary in the world
each county boundary in the US (thousands)
other countries' locality boundaries
city boundaries
TNC Terrestrial Ecoregions http://maps.tnc.org/files/shp/terr-ecoregions-TNC.zip
Census shapefiles https://www.census.gov/geographies/mapping-files/time-series/geo/carto-boundary-file.html
US state ecoregion files https://www.epa.gov/eco-research/ecoregion-download-files-state-region-4

jonhealy1 commented 2 years ago

Should we be doing some CRUD stuff at the same time to simulate a service that is adding data every day potentially? It could be optional.

philvarner commented 2 years ago

I just added four initial test:

Query each geometry in the STEP database
Query each geometry in the TNC Terrestrial Ecoregions database
Query one item repeatedly
Query locations that have no results in the AOI