Many Databases 1 LSM Engine - OpenData

Modern database infrastructure suffers from an excessive "operational tax" caused by redundant distributed systems plumbing across various engines. Almog Gavra, co-founder of Responsive, argues that shifting to an object storage-native foundation—specifically using an LSM tree architecture like SlateDB—can consolidate these systems. By utilizing object storage as the primary source of durability, developers can abstract away complex replication and fencing protocols, allowing diverse databases like vector, time series, and log stores to share a common storage layer. This model enables independent scaling of readers and writers and provides unified tuning knobs for read, write, and space amplification. Ultimately, this approach prioritizes reducing the human and operational burden of database management over merely optimizing hardware costs, offering a more efficient, simplified framework for the majority of data-intensive applications.