DBDB.io The Encyclopedia of Database Systems · Est. 2017
Database of Databases

Database Entry

HydrAIDE


HydrAIDE is a data engine designed for workloads that scale by namespace. It organizes data into independent namespaces called Swamps, which reside in separate files on disk and load into memory only when accessed. Data within a Swamp is stored as typed key-value pairs referred to as Treasures. The system uses a gRPC wire protocol and provides a Go SDK for direct struct serialization. Each write operation emits a real-time event via a Subscribe stream.[04]

Source Code
https://github.com/hydraide/hydraide[02]
Developer
Country of Origin
HU
Start Year
2025
Coding Agents
Project Type
Open Source
Written in
Go
Supported Languages
Go
License
Apache v2

Database Entry

HydrAIDE


HydrAIDE is a data engine designed for workloads that scale by namespace. It organizes data into independent namespaces called Swamps, which reside in separate files on disk and load into memory only when accessed. Data within a Swamp is stored as typed key-value pairs referred to as Treasures. The system uses a gRPC wire protocol and provides a Go SDK for direct struct serialization. Each write operation emits a real-time event via a Subscribe stream.[04]

History


Development began in 2021 to address storage limitations encountered while indexing millions of European websites for the Trendizz project. Existing relational and document databases, as well as disk-based B-tree engines, did not meet the required latency, memory efficiency, or cost constraints for handling tens of millions of independent storage units on a single server. The system has been used in production for Trendizz.com since 2024.

Coordination


Data Model


Foreign Keys


Indexes


Joins


Query Interface


Storage Format


Stored Procedures


System Architecture


Views


Revision #3