Multiple ChoiceHard 45s
LSM tree read amplification tradeoff
Why do LSM-tree storage engines typically suffer higher read amplification than B-trees?
Options
- 1Because keys are stored uncompressed in each SSTable, so a read must decompress every block it touches before comparing
- 2Because a read may need to probe multiple SSTables and levels, plus bloom filters, before finding the key
- 3Because LSM trees cannot maintain secondary indexes, so every point lookup falls back to a full linear scan of the engine
- 4Because writes are batched in an in-memory table, so a read waits for background flushes before it can locate a committed key
Think you can answer under pressure?
Get matched and answer live questions in 3-minute 1v1 battles. Elo ranking, badge tiers and bragging rights on the line.