Here, a cluster of commodity hardware scales to 8.2 million operations/sec doing a 90% read and 10% write workload.
Make sure to check out our blog post about scaling a cluster that reaches 14.4 million writes/second.+ More about scaling
The scaling graph uses a 24-machine EC2 c3.8xlarge cluster in which each machine has a 16-core processor. We ran a Key-Value Store server process on each core, yielding a 384-process cluster for the largest test, and scaled the cluster down for each smaller test.
Scaling is the ability to efficiently deliver operations at different scales. For Key-Value Store, the relevant operations are reads and writes, measured in operations per sec. Scale is measured in the number of processes, which will usually track the number of available cores. Key-Value Store offers scalability from partial utilization of a single core on a single machine to full utilization of dozens of powerful multi-core machines in a cluster.