EloqKV, a distributed database with Redis compatible API (GPLv2 and AGPLv3)

Summary

EloqKV EloqKV is a high-performance distributed database with a Redis/ValKey compatible API. It offers features like ACID transactions, full elasticity and scalability, tiered storage, and session-style transaction syntax — all while preserving Redis' simplicity and usability. EloqKV is engineered for developers who need a modern no-compromise database solution to power the next generation of demanding applications in the AI era. Why Choose EloqKV Over Redis? Feature Redis EloqKV High Performance Single-threaded Multi-threaded (1.6million QPS on c6g.8xlarge) Transactions MULTI/EXEC (No Rollback) Redis API plus BEGIN/COMMIT/ROLLBACK (ACID) Distributed Transactions CROSSSLOT Error ACID distributed transactions Data Durability Limited, AOF/RDB snapshots Replicated WAL + Tiered Storage Cold Data Must fit in memory Auto-tiering to disk Client Transparency Cluster needs specific client Same client for a single server or a cluster Key Features ⚡ High Performance Multi-threaded : Built with thread-per-core execution and message-passing architecture to fully utilize modern multicore CPUs. : Built with thread-per-core execution and message-passing architecture to fully utilize modern multicore CPUs. Single Node : Up to 1.6M QPS on AWS c6g.8xlarge, comparable to purpose-built cache systems like DragonflyDB and far out-performs Redis and Valkey. : Up to on AWS c6g.8xlarge, comparable to purpose-built cache systems like DragonflyDB and far out-performs Redis and Valkey. Natively Distributed: Scale horizontally with distributed transactions, so your application works the same whether it's backed by a single-node EloqKV or a cluster of servers. 🗃️ Full Durability with Tiered Storage WAL for True Durability : No more data loss due to power failures. : No more data loss due to power failures. Hot Data : In-memory for microsecond access. : In-memory for microsecond access. Cold Data: Automatically offloaded to disk. Save 70% on memory costs compared to pure in-memory cache such as Re...