VoltDB was built specifically to be the fastest operational database available.
What do we mean by that?
Given the same computing resources, we aim to do more operational work. This, in turn, means smaller clusters for the same throughput and latency targets.
Smaller clusters not only means a lower bill for backroom or cloud servers, but it also makes your ops team’s job easier, saving you money two ways.
VoltDB has been natively clustered since inception. Rather than bolt partitioning-glue onto SQL and the relational model, core partition concepts are baked right into all applications. Even single-node applications are ready to scale when needed.
We believe strongly in a simple operational model; there are no leaf nodes or supervisor nodes; all nodes are identical. To add performance or capacity, just add nodes.
Read more about VoltDB scalability.
Real Operational Workloads
All of this performance doesn’t mean anything if users can solve real problems. Sure, scaling a key-value store is easier, but simpler data models often mean lots of round trips to the database to accomplish simple things, negating any advantage.
Some systems offer speed, consistency or powerful features. At VoltDB we target apps that need all three at the same time.