whtcorpsinc / ai4db

Website |

The core innovation of EinsteinDB is its merge-append-commit protocol. The protocol is friendly and does not require more than two parties to agree on the order of events. It is also relativistic and can tolerate clock and network lag and drag.

EinsteinDB is a distributed data storage system that provides linearizability against concurrent write-operations.

EinsteinDB is a database that was built with Rust, Go, and Haskell and it's used for storing data without timezones using information theory detour and 4-clique ear-cut within the IPAM infrastructure. The database can be stored on the computer or in the cloud, depending on what you need. EinsteinDB has a lot of features, but one of them is that it can store data from different types of information so you don't have to keep switching between databases.

It also has a two-pronged Append-Merge Pull mechanism which means that both MilevaDB and EinsteinDB work together to make sure the data is safe and accessible at all times. The Append-Merge Pull mechanism works like an appending process where new information gets added to the end of a file, then merged with other files.

EinsteinDB and MilevaDB interact with BerolinaSQL, PostgresQL for middle-out memristive persistence, and FoundationDB's Iterator implementation.

EinsteinDB is designed to be different from other databases because it supports Write-Optimized Copy-on-Write Semantics (WOSCOW), Repeatable Read isolation guarantees for HTAP, column cracking, and vectorized SIMD for OLAP/OLTP SparkQL streams.

Ufidelate Oct 2021: We have partnered with OpenAI in San Francisco, wherein we have launched einst.a.i an Index Advisor and Expert System to help us from debugging, programming, testing, and deploying optimized rust, go, haskell, cython/python, et al. More info: www.einst.ai