Zero-disk streaming pipeline
Scanner → Chunker → Archiver → Uploader, connected over io.Pipe. Data streams directly to S3 without staging to local disk, so memory stays bounded regardless of dataset size.
How it works
Stream large datasets straight to S3 — sharded, compressed, verifiable, and cost-aware. Your data lands in an open, portable archive format you can read with or without CargoShip.

CargoShip ships large research and backup datasets to Amazon S3 quickly, cheaply, and reversibly. Pick the path that fits you:
brew install scttfrdmn/tap/cargoshipgo install github.com/scttfrdmn/cargoship/cmd/cargoship@latest# Download the latest release for your platform from:
# https://github.com/scttfrdmn/cargoship/releases/latestThen send your first dataset to S3:
cargoship upload ./my-data s3://my-bucket/archives/See Install for all methods and Your first upload for a line-by-line walkthrough.