EloqKV: Achieving Predictable P99.99 Latency on NVMe with Redis API

Re:EloqKV: Achieving Predictable P99.99 Latency on NVMe with... by the_precipitate on Friday January 23, 2026 @11:14AM (#46734233)

With DRAM price this high, this is certainly a welcome feature. But how do you control write latency? B+ Tree is pretty bad at updates and LMDB, another BTree based storage is lightning fast on reads but quite bad on writes compared with RocksDB.

[ Parent | on HN ]

Re:EloqKV: Achieving Predictable P99.99 Latency on NVMe with... by iamlintaoz on Friday January 23, 2026 @11:21AM (#46734327)

The disk storage EloqKV uses (EloqStore [1]) is optimized for batch updates because the upper Data Substrate layer manages buffering and the Write-Ahead Log (WAL), absorbing writes and guaranteeing durability. When durability is not required, the WAL can be optionally disabled.

[1] github.com/eloqdata/eloqstore

Disclaimer: I am the CEO of EloqData

[ Parent | on HN ]

Re:EloqKV: Achieving Predictable P99.99 Latency on NVMe with... by hubertzhang on Friday January 23, 2026 @11:32AM (#46734460)

we leverage batch write optimization which uses Copy-on-write B-tree variant enables high-throughput batch writes without blocking concurrent reads. MVCC-based design eliminates lock contention and provides predictable write amplification.

[ Parent | on HN ]