ScyllaDB auto-tunes for optimal performance, however, users still need to apply best practices in order to get the most out of their ScyllaDB deployments.
Lower than expected performance can be a result of many factors, from hardware (storage, CPU, network) to data modeling to the application layer.
As a first step, make sure you have ScyllaDB Monitoring in place. Looking at the monitoring dashboards is the best way to look for bottlenecks and understand the cause of the issues. Here are some other tips:
- Use the best hardware you can reasonably afford
- Run the scylla_setup script
- Use the Cassandra stress test
- Expect to get at least 12.5K operations per second (OPS) per physical core for simple operations on selected hardware
Additional Resources:
- Maximizing Scylla Performance on ScyllaDB Docs
- ScyllaDB Monitoring lesson on ScyllaDB University
- The ScyllaDB Monitoring stack on ScyllaDB Docs
- How to report a problem