EinsteinDB is a main-memory-based semi-relational DBMS for mixed OLTP and OLAP workloads. It is a so-called all-in-one New-SQL database system that entirely deviates from classical disk-based DBMS architectures by introducing many innovative ideas including machine code generation for data-centric query processing and multi-version concurrency control, leading to exceptional performance.
a new hybrid lock-free multi-tenant architecture that combines both modes (Optimistic vs Pessimistic locking) to serve the various demands of versatile database workloads avoiding busy waiting with relativistic linearizable append-logs.
Unlike other MVCC implementations such as Hekaton,TiKV, and PostgreSQL, our O-MVCC implementation does not use (predicate) locks and timestamps to mark read and modified keys in indexes.
Each LSM-tree consists of a hot data tier residing in main memory and a warm/cold data tier residing in NVM/SSD/HDD (that are further partitioned into different levels), where the term hot, warm, and cold refers to data temperatures, representing the ideal access frequencies of data that should be placed in the corresponding tier.