CompanyJanuary 29, 2020

Introducing DSBench

Performance and Scale Testing, Data Model Validation and More For Apache Cassandra™ and DataStax Enterprise
Jonathan Shook
Jonathan Shook
Introducing DSBench

DataStax is launching our first release of DSBench in DataStax Labs.

Streamline model validation, performance and scale testing for Cassandra and DSE

With DSBench, you can run performance tests against DataStax Enterprise (DSE) or Apache Cassandra, validate data models, bring scale testing into your continuous integration practices, and more. You can use it anywhere you need to put a realistic load on a target cluster to characterize performance. Also, It’s free!

DSBench is Different

We wanted a testing toolkit that could be used by everyone, with a common set of concepts and a standard configuration format. DSBench provides this and much more; its machinery is different from other performance testing tools. We designed DSBench to help our users understand their systems with minimal effort.

We’re sharing a working version of DSBench that is still improving. We want to start getting feedback about what works well, what doesn't, and what enhancements will help our users the most.

DSBench is Simple

This means that if you want to do something simple, you should be able to quickly find the right command or recipe and just do it. It also means that if you need to do something more nuanced or sophisticated, it should be possible with a reasonable investment of effort. 

DSBench is Powerful

DSBench speaks the language of workloads — data, operations and access patterns. It can make your performance testing relevant and specific to your requirements. It doesn’t require compilation when you need to make changes to a workload. It doesn’t require you to manage large amounts of test data to get realistic results. It emulates the access patterns of your application. It has built-in instrumentation that can plug into most metrics systems. DSBench is fast and efficient. 

Simple AND Powerful?

Testing distributed systems performance is not easy. We hope this toolkit grows up to be the de-facto testing tool for NoSQL systems. To make this possible, it is essential that we allow users to start at the top and drill down as far as they need. This is another way of saying “simple things should be simple; complex things should be possible,” and it is one of the philosophical cornerstones of the designs behind DSBench.

By keeping the simple things simple, we can enable better composability of the parts into a greater whole. If you’re a casual user who just wants a ballpark baseline of a system, there are example commands waiting for you to copy and paste. If you are a test engineer who is responsible for thorough performance and capacity planning on a new or existing service, then you will find advanced functionality and examples that can be adjusted to your testing requirements.

We’re Just Getting Started

With this release, we’re just getting the ball rolling. There will be many new and interesting things to share as we go, as well as lessons to learn along the way.

DataStax is providing places for us to ask questions, manage bugs, and collaborate in general on bringing DSBench to everyone. Please join us in this new endeavor — try it out, and let us know what you think!

To get started, head over to DataStax Labs to get to the DSBench Getting Started guide.

Discover more
DataStax EnterpriseApache Cassandra™

One-stop Data API for Production GenAI

Astra DB gives JavaScript developers a complete data API and out-of-the-box integrations that make it easier to build production RAG apps with high relevancy and low latency.