Financial market data
Tick archives you can actually query — 18.7× smaller, and half the size of Parquet
AT-1 stores exchange ticks 18.7× smaller, byte-for-byte lossless, and a time-and-symbol query reads ~1% of the file (54× less I/O) returning the exact rows — the same archive serves storage and query.
Who it's for: For quant desks, exchanges, and market-data teams whose tick/quote archives grow without bound.
18.7×
Binance ticks
1.49×
vs xz-9
54×
less query I/O
½
the size of Parquet
How it fits: Land your tick CSV/Parquet into AT-1 (qcolumnar); query by time range and symbol over object storage with range-GET — no warehouse to spin up, and the original is always recoverable.
Real Binance BTCUSDT aggTrades; queryable-while-compressed measured in bench_query.
Put a number on your data.
Enter your storage bill; see the saving and a one-page proposal in 30 seconds.