What you get.
Placeholder: a few lines on how scientific formats works and why it matters.
- Drop-in for existing pipelines
- Chunked for cloud reads
- Read a slice without decoding the file
- Metadata kept intact
Used for
Where scientific formats helps.
Network telemetry
Keep every counter and probe record, not a sample.
See howBroadcast archives
Decades of programmes, smaller and still broadcast-grade.
See howPost-production
Camera originals that don’t fill the SAN.
See howSpectrum monitoring
Record the whole band, not just what you flagged.
See howRadio astronomy
Telescope data that outruns any disk.
See howObservability logs
Keep every line, without the bill.
See howData lakes & analytics
Event and analytics tables, a fraction of the size.
See howDrone mapping & survey
Upload a day’s flights before the next one.
See howISR full-motion video
More eyes on target through the same pipe.
See howPublic records & archives
Digitised records, kept forever for less.
See howGeospatial & mapping agencies
National imagery and elevation data, smaller.
See howClimate & weather archives
Petabytes of model output in a fraction of the space.
See howForecast distribution
Get each model run to users before the next one lands.
See howWeather radar
Every sweep from every radar, kept.
See howCrop imagery & remote sensing
Season-long imagery for every paddock.
See howSimulation & HPC
Checkpoints and outputs that no longer fill the cluster.
See howSeismic & geophysical sensors
Dense arrays, recorded continuously.
See howFeatures
More features.
Neural codecs
Codecs trained on your kind of data.
Learn moreError bounds
Set the error per variable; we guarantee it.
Learn moreEdge & streaming
Compress at the sensor, before the link.
Learn moreSDK & CLI
One line in Python, a command in the shell.
Learn moreCloud, on-premises or air-gapped
Runs where your data already lives.
Learn moreWe’ll cover
- Compress run on a sample of your data
- Size, speed and error compared with what you use now
- Fitting it into your pipeline
- Deployment: cloud, on-premises or edge