I like how ScyllaDB makes large-scale distributed data feel simple and allows me to work with its design patterns effortlessly. I really appreciate the automatic distribution and the hot reload performance of 1 million operations per second. One thing I especially like about ScyllaDB is its high performance and predictable low latency, even under heavy workloads. I also value its compatibility with the Cassandra ecosystem, which makes migration and integration straightforward. Features like shard-aware drivers, horizontal scalability, and efficient resource utilization work particularly well for large-scale, high-throughput applications like ours. Review collected by and hosted on G2.com.

Overall, my experience with ScyllaDB has been positive, but there are a few areas that could be improved. Initial cluster setup and tuning can be complex, especially for production deployments. Troubleshooting shard-awareness and networking issues also required a good understanding of the architecture. I also think the documentation could include more end-to-end examples and performance tuning guides for large-scale deployments. I would like to see more built-in observability and performance diagnostics. Additionally, although I use ScyllaDB with GUI tools like DataGrip and DBeaver for querying data, browsing tables, and schema management, I feel these tools are often paid and limit users for a better experience interacting with ScyllaDB. Review collected by and hosted on G2.com.