Data Compression
Data compression reduces the size of stored data using algorithms that encode it more efficiently. In databases it cuts storage cost and often speeds up queries, because less data has to be read from disk.
Why compression speeds up analytics
Compression is usually framed as a way to save storage space, and it does. But in analytics it also makes queries faster. The slow part of a big scan is often reading data off disk. If the data is compressed to a fraction of its size, there is far less to read, so the query finishes sooner even after accounting for the work to decompress.
Columnar data compresses especially well. Because each column holds similar values, encoders find a lot of repetition to exploit. Algorithms like ZSTD can shrink columnar data dramatically with fast decompression.
The result is the rare win where you save money and gain speed at the same time.
How Arc handles Data Compression
Arc stores data as ZSTD-compressed Parquet, typically around 90 percent smaller than raw. That cuts storage cost and helps queries run fast, since the engine reads far less data off object storage.
Arc is a high-performance columnar database. Open Parquet on storage you own, single Go binary, production-ready in 30 seconds.