ScyllaDB Enterprise

What do you like best about the product?

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.

What do you dislike about the product?

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.

What problems is the product solving and how is that benefiting you?

I use ScyllaDB for managing large-scale distributed storage, automatic distribution, and high throughput reloads without cluster degradation. It handles even data distribution, sub-millisecond latencies on read-heavy workloads, and offers compatibility with Cassandra for easy migration, enhancing our scalability and resource utilization.

What do you like best about the product?

ScyllaDB provides exceptional throughput and sub-millisecond, predictable latency. Because it is written in C++ rather than Java, we completely eliminated garbage collection pause issues and saw a massive performance boost over Cassandra with far fewer nodes.

What do you dislike about the product?

ScyllaDB delivers high performance, but its ecosystem of third-party monitoring plugins and integrations is still a bit smaller than what you’ll find with older databases such as Apache Cassandra.

What problems is the product solving and how is that benefiting you?

Before we started using ScyllaDB for our distributed systems coursework and hackathon prototypes, we often ran into severe latency spikes and unpredictable Garbage Collection (GC) pauses with traditional Java-based NoSQL databases such as Cassandra, especially under high concurrent loads. ScyllaDB removed that bottleneck by leveraging its C++ architecture along with a shared-nothing, shard-per-core design.

By avoiding JVM overhead, our applications can keep sub-millisecond, predictable latency even when traffic suddenly spikes. And as students working with minimal cloud resources and strict hardware budgets, we’ve found that ScyllaDB delivers much higher throughput on a significantly smaller footprint, helping us maximize single-node CPU and RAM utilization without needing expensive multi-node clusters.

What do you like best about the product?

Ease of use, Ease of Implementation, number of features

What do you dislike about the product?

nothing sepesific yet still assessing, did not try yet Customer Support

What problems is the product solving and how is that benefiting you?

did not decide yet on how to use it and for what; still evaluating

What do you like best about the product?

The integration with Kafka offers low latency and high throughput.

What do you dislike about the product?

Nowadays, I don’t dislike it at all; for me, it’s the best database in the world.

What problems is the product solving and how is that benefiting you?

It helps us solve real-time event problems. We used to work only with Kafka, but now we can work with ScyllaDB too.

What do you like best about the product?

I like ScyllaDB for its impressive performance and speed, offering 10 times more performance compared to Cassandra. It's rewritten in C++ instead of Java, which provides predictable low latency, like p99 latencies under heavy load, and overcomes JVM limitations. I also appreciate its ability to tune settings, reducing administrative overheads and making it cost-efficient as well.

What do you dislike about the product?

It could be the steep learning curve. You know, optimizing and, managing the ScyllaDB at scale, it requires deep knowledge of distributed systems. Especially around the horizontal scaling using the shard key selections and data modeling. Sometimes it warrants a complex data modeling, because this is crucial. To avoid those expensive mistakes at a later stage. The third one would be a, smaller community. So even though it is growing, but there is still a smaller community, compared to the Postgres and other databases available.

What problems is the product solving and how is that benefiting you?

I use ScyllaDB for its flexible schema and horizontal scaling, solving our high-performance application needs with low latency, particularly during heavy loads like 'Black Friday'. Its performance outpaces peers, offering cost efficiency by eliminating administrative overheads.