Distributed Databases: Types, Advantages, and 10 Popular Implementations

In this article the author, after previously discussing MySQL versus NoSQL, introduces the topic of databases in distributed scenarios, emphasizing the need for high performance and reliability when scaling beyond a single machine.

1. Relational vs. Non‑relational – Relational databases use the 关系模型 (tabular model) and include systems like MySQL, Oracle, PostgreSQL, while NoSQL databases (key‑value, column‑family, document, graph) such as Redis, HBase, MongoDB, Neo4j offer flexible schemas and horizontal scalability. The article lists the advantages (SQL support, transaction safety, data integrity) and drawbacks (schema rigidity, lock‑table issues, high memory usage) of relational databases, and the strengths (flexibility, speed, low cost) and weaknesses (lack of SQL, limited transaction support) of NoSQL.

2. Distributed Databases – A distributed database is defined as a database implemented with a distributed architecture. Its main advantages are improved 性能 (performance) and 可靠性 (reliability) through multi‑node replication and fault tolerance. The author notes that major companies like Alibaba (OceanBase), Tencent, ByteDance, Meituan, and many banks have adopted distributed databases.

2.3 Characteristics – The article distinguishes OLTP (online transaction processing) and OLAP (online analytical processing) workloads, highlighting that distributed databases excel in write‑heavy, low‑latency, high‑concurrency OLTP scenarios, offering features such as strong consistency, massive storage, and high availability.

3. Ten Distributed Database Solutions

3.1 TiDB – An open‑source HTAP database with horizontal scaling, high availability, MySQL compatibility, and components TiDB (SQL engine), TiKV (key‑value store), PD (metadata).

3.2 Google Spanner – A globally distributed, multi‑version database providing external consistency via Paxos , Tablet sharding, and 2PC transaction management.

3.3 CockroachDB – A scalable SQL database using the Raft consensus algorithm, offering strong consistency, survivability, and global deployment.

3.4 YugabyteDB – Similar to CockroachDB, with PostgreSQL compatibility, hybrid logical clocks, and a two‑layer architecture (query layer and storage layer).

3.5 Alibaba OceanBase – A financially‑grade distributed relational database with high consistency, high availability, and strong performance.

3.6 Tencent TDSQL – A MySQL‑based distributed cluster supporting flexible deployment and mixed workloads.

3.7 ZTE GoldenDB – A large‑scale distributed database used in banking, based on MySQL with a global clock node.

3.8 Tencent TBase – An enterprise HTAP database built on PostgreSQL, offering distributed transactions, high performance, and comprehensive security features.

3.9 VoltDB – An in‑memory relational database delivering 100× faster query speed, strong consistency, and horizontal scalability.

3.10 SequoiaDB – An open‑source financial‑grade distributed relational database supporting multiple instance types (MySQL, PostgreSQL, MongoDB‑like, S3, POSIX).

The article concludes by inviting readers to share their own distributed database experiences and provides a list of reference links.