During RethinkConn 2022, Jean-Noel Moyne, Synadia’s Field CTO, presented a lightning talk Comparing & Contrasting NATS to Kafka. This has been the most popular video on the event’s YouTube playlist and led to a growing number of requests from our community for more content on this topic.
The consistent feedback we receive from customers and the community is the appreciation for NATS overall simplicity. This manifests in the ease and power of developing NATS applications as well as the operability of a NATS deployment.
One of the most repeated requests we receive has been for a total cost of ownership (TCO) comparison against several real-world, and comparable workloads. Users want to know if the “simplicity” is a trade-off vs functionality or performance.
Synadia engaged and sponsored the Mcknight Consulting Group to conduct an independent evaluation of this comparison.
We were delighted that the conclusions from the TCO analysis validate that NATS gives you both simplicity and performance… and it will cost you less 🙂. Check out the full report for the details of testing methodology, benchmarks, and analysis.
That said, the NATS maintainers' stance has always been to evaluate the performance for your specific use case and requirements. This was the motivation for building a robust benchmarking capability directly into the NATS CLI to make it simple to model and benchmark workloads against your real world environment.
As highlighted in the presentation, there are fundamental architectural differences between the two platforms. A white paper discussing the functional and architectural differences between NATS and Kafka is going to be released in a subsequent post.
If you are just getting started with NATS, take advantage of the various resources that are available:
Documentation - https://docs.nats.io
Examples - https://natsbyexample.com/
Screencasts - https://rethink.synadia.com/
Newsletter - https://synadia.com/newsletter
Slack - https://slack.nats.io
