What RocksDB does

RocksDB is an embeddable, persistent key-value store library originally developed in 2013 by Facebook's Database Engineering Team (now Meta Platforms, Inc.), based on LevelDB by Google. It implements a log-structured merge-tree (LSM-tree) engine written entirely in C++ and is optimized for fast, low-latency storage environments such as flash drives, high-speed disks, and RAM. The library supports arbitrarily-sized byte-stream keys and values ordered by a user-specified comparator, and exposes basic operations (Put, Get, Delete) alongside advanced capabilities including compaction filters, merge operators, multiple filter types (Bloom and Ribbon), compression dictionaries (zstd, LZ4, zlib), tiered storage with hot/warm/cold placement, async IO via C++ coroutines, unified memory tracking, and per-key-value xxh3 checksums. APIs are provided in C++ and Java, the latter with both JNI and a prototype Foreign Function Interface path.

Name

RocksDB

Legal name

Meta Platforms, Inc.

Website

https://rocksdb.org

Company type

Private

Founded year

2013

Operating status

Operating

Headcount range

1–10 employees

Short description

RocksDB is an open-source embeddable persistent key-value store library written in C++ with an LSM-tree engine, maintained by Meta's Database Engineering Team and used by developers building database storage engines, application caches, and embedded workloads.

Ownership category

akta.pro rank

Product category

Embedded Key-Value Store

NAICS

Computer Systems Design and Related Services (54151), Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (51821)

SIC

Services-Prepackaged Software (7372), Services-Computer Programming, Data Processing, Etc. (7370), Computer Storage Devices (3572)

akta.pro primary industry

Key-Value Stores (HDAEAAAH)

Keywords

• Embedded key-value store
• Persistent storage engine
• LSM-tree database
• C++ storage library
• Open-source database

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel

Revenue model

1. Open Source Distribution: RocksDB is an open-source project maintained by Meta Platforms, Inc. It is freely available under a BSD-style license. There is no commercial revenue model; the project is maintained as internal infrastructure that Meta open-sources for community benefit.

Distribution channels2 records

Marketing channels6 records

Core offering

RocksDB is an embeddable persistent key-value store library written in C++ and based on Google's LevelDB. It is optimized for fast, low-latency storage environments such as flash drives and high-speed disk drives, using an LSM-tree architecture to deliver high read/write throughput. The library provides basic operations (Put, Get, Delete) and advanced operations (merging, compaction filters) with arbitrary byte stream keys and values, and is distributed as open-source software under a BSD-style license.

Product overview

RocksDB is a single embeddable persistent key-value store product. It is a C++ library that provides a persistent key-value store for fast storage environments. The core product offers high-performance key-value storage with APIs in C++ and Java. RocksDB uses a log-structured database engine with support for basic operations (Put, Get, Delete) and advanced operations (merging, compaction filters). The library is maintained by the Facebook Database Engineering Team and is based on LevelDB by Google.

Differentiator

Problem solved

Functional benefit

Brands

• Meta Open Source: The initiative under which Meta releases open-source projects including RocksDB

Products and services

• RocksDB An embeddable persistent key-value store optimized for fast storage environments such as flash drives and high-speed disk drives. It uses a log-structured database engine written entirely in C++ for maximum performance, with keys and values as arbitrarily-sized byte streams ordered by a user-specified comparator function. The library provides APIs in C++ and Java and is intended for developers building database storage engines (such as MyRocks), application data caching layers, and embedded workloads.

Quantifiable outcome

• Range tombstone conversion achieves ~99x speedup in forward scans and ~368x in reverse scans for delete-heavy workloads
• +5 more outcomes

Named customers4 records

Segments3 records

Ideal customer profiles2 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Feature15 records

Partnerships

One partnership is on record.

• Google (LevelDB)foundationalTechnology or IntegrationRocksDB is based on LevelDB, an open-source key-value store developed by Sanjay Ghemawat and Jeff Dean at Google. LevelDB provides the foundational architecture that RocksDB builds upon and extends with additional features and optimizations for production use at scale.

Scale indicators2 records

Recent moves6 records

Expansion highlights4 records

Direct peers

• etcd-io (CNCF): Maintainer of BoltDB (bbolt), an embedded key-value store used by etcd. While B+tree-based rather than LSM, it directly competes with RocksDB in embedded persistence scenarios for cloud-native and distributed systems.
• Dgraph Labs: Creator of BadgerDB, an embeddable persistent key-value store written in Go using LSM-tree architecture. BadgerDB directly competes with RocksDB in embedded KV scenarios, particularly in Go-native environments.
• Symas: Maintainer of LMDB, a high-performance embedded transactional key-value store using B+tree architecture. LMDB is a long-standing alternative to RocksDB in embedded and directory-backed storage scenarios.

Emerging players

• ScyllaDB: High-performance NoSQL database that uses RocksDB as its underlying storage engine. Comparable as a downstream consumer and as a competing purpose-built data store targeting similar low-latency workloads.
• InfluxData: Time-series database that uses RocksDB as its storage engine for local persistent storage. Comparable as a downstream RocksDB adopter and as a competing data platform for time-series workloads.
• PingCAP (TiKV): Developer of TiKV, a distributed transactional key-value database that uses RocksDB as its underlying storage engine. Comparable through its deep integration with RocksDB and focus on cloud-native OLTP workloads.

Broad incumbents

• MongoDB: Uses RocksDB as one of its storage engine options (alongside WiredTiger). MongoDB is a major consumer of RocksDB and a relevant comparator as a database platform that integrates embedded storage engines.
• Google (LevelDB): Author of LevelDB, the foundational library on which RocksDB is built. While maintained at lower velocity than RocksDB, LevelDB remains a direct LSM-based embeddable alternative for simpler workloads.
• Aerospike: Real-time NoSQL database with hybrid memory/flash storage targeting low-latency key-value workloads. Competes with RocksDB-based systems in the high-throughput time-series and ad-tech segments.
• Apple (FoundationDB): Distributed key-value store underpinning several Apple cloud services. Comparable as an alternative distributed/embedded KV design with a focus on consistency and fault tolerance rather than raw LSM throughput.

Market position

Strengths4 records

Weaknesses5 records

Competitive moat5 records

Key risks5 records

Key highlights6 records

Customer concentration

• Xx.com/RocksDB
• GitHubgithub.com/facebook/rocksdb
• Facebookfacebook.com/rocksdb

Revenue estimate

Valuation estimate

Number of profiles

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about RocksDB