Columnar serialization allows a database to scan only the columns required for a particular query, sometimes dramatically reducing the amount of data read from the disk. In addition, arranging data by column packs similar values next to each other, yielding high-compression ratios with minimal compression overhead. This allows data to be scanned more quickly from disk and over a network.
Columnar databases perform poorly for transactional use cases—i.e., when we try to look up large numbers of individual rows asynchronously. However, they perform extremely well when large quantities of data must be scanned—e.g., for complex data transformations, aggregations, statistical calculations, or evaluation of complex conditions on large datasets.