Understanding MySQL Cluster: Architecture, Scalability, and 200 Million QPS Benchmark
The article explains MySQL Cluster as a real‑time, ACID‑compliant, in‑memory distributed database, describes its node architecture, transparent sharding for linear scalability, various SQL/NoSQL access methods, and presents a benchmark achieving 200 million queries per second on 32 data nodes.
Andrew Morgan, Oracle MySQL’s chief product manager, introduces MySQL Cluster, a real‑time, ACID‑compliant, in‑memory distributed database offering up to 99.999% availability and low total cost of ownership.
MySQL Cluster Overview
Designed originally for telecom applications requiring carrier‑grade availability and real‑time performance, MySQL Cluster now serves web, mobile, and enterprise workloads such as large‑scale OLTP, real‑time analytics, e‑commerce, online gaming, fraud detection, and more.
MySQL Cluster Architecture
Three node types provide services: Data Nodes (store data in memory and optionally on disk, handle sharding, replication, transactions, and automatic failover), Application Nodes (connect applications via standard MySQL connectors, NDB API, or NoSQL interfaces like Java, JPA, Memcached, JavaScript/Node.js, HTTP/REST), and Management Nodes (distribute configuration, manage cluster membership, and provide arbitration).
Scalability Through Transparent Sharding
Each table row can be split into multiple shards, each stored on a dedicated Data Node with a partner node holding a replica. The cluster uses a two‑phase commit protocol to ensure synchronous replication.
Shard keys default to the primary key, hashed with MD5 to determine placement. Transactions spanning multiple shards are coordinated by one node, with results merged before returning to the client.
Maximizing Data Access Speed with NoSQL APIs
In addition to standard SQL, MySQL Cluster offers native APIs for C++, Java, JPA, JavaScript/Node.js, HTTP, and Memcached, allowing applications to bypass the SQL layer for lower latency.
Benchmark: 200 Million Queries per Second
The cluster targets two workloads: OLTP (memory‑optimized tables delivering sub‑millisecond latency) and real‑time search (high concurrency scans of non‑indexed columns). Using the flexAsynch benchmark on MySQL Cluster 7.4, throughput scales linearly up to 32 Data Nodes, reaching 200 M NoSQL queries per second.
For detailed results, visit the MySQL Cluster benchmark page and the MySQL Cluster 7.4 release announcement.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Qunar Tech Salon
Qunar Tech Salon is a learning and exchange platform for Qunar engineers and industry peers. We share cutting-edge technology trends and topics, providing a free platform for mid-to-senior technical professionals to exchange and learn.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.