What are the typical performance metrics for MySQL Cluster?
• Availability • 99.999% ( • Performance • Response Time 2-5 millisecond (with synchronous replication and access via NDB API) • Throughput of 10,000+ replicated transactions/sec on a 2 Node Cluster, with 1 CPU Per Node (minimal configuration) • Throughput of 100,000 replicated transactions/sec on 4 Node Cluster, with 2 CPU Per Node (low-end configuration) • Failover • Sub-second failover enables you to deliver service without interruption • Scalability • Scale out, scale up and scale dynamically. • For cost-effective scale-out: • Add more data nodes (8, 10, 12,…48) per cluster, or • Add more CPUs (4, 8, etc.) or • Add more Memory (16GB, 32GB, etc.) per data node MySQL Cluster offers a distributed, in-memory database* so for atomic requests (like primary key lookup) it is very fast and predictable, but the distributed architecture is not ideal for very complex queries with lots of joins. The best performance can be achieved if the application has distribution awareness (allowing all
• Availability • 99.999% ( • Performance • Response Time 5-10 millisecond (with synchronous replication) • Throughput of 10,000+ replicated transactions/sec on a 2 Node Cluster, with 1 CPU Per Node (minimal configuration) • Throughput of 100,000 replicated transactions/sec on 4 Node Cluster, with 2 CPU Per Node (low-end configuration) • Failover • Sub-second failover enables you to deliver service without interruption • Scalability • Near-linear scalability on preferred system configurations. • For cost-effective scale-out: • Add more storage nodes (8, 16, 32, etc.) per cluster, or • Add more CPUs (4, 8, etc.) or • Add more Memory (16GB, 32GB, etc.
• Availability • 99.999% ( • Performance • Response Time 2-5 millisecond (with synchronous replication and access via NDB API) • Throughput of 10,000+ replicated transactions/sec on a 2 Node Cluster, with 1 CPU Per Node (minimal configuration) • Throughput of 100,000 replicated transactions/sec on 4 Node Cluster, with 2 CPU Per Node (low-end configuration) • Failover • Sub-second failover enables you to deliver service without interruption • Scalability • Scale out, scale up and scale dynamically. • For cost-effective scale-out: • Add more data nodes (8, 10, 12,…48) per cluster, or • Add more CPUs (4, 8, etc.) or • Add more Memory (16GB, 32GB, etc.) per data node MySQL Cluster offers a distributed, in-memory database* so for atomic requests (like primary key lookup) it is very fast and predictable, but the distributed architecture is not ideal for very complex queries with lots of joins. The best performance can be achieved if the application has distribution awareness (allowing all
