Start Now

Wrapping Up: Jepsen Test Results for YugabyteDB 1.2 Webinar

Jimmy Guerrero

VP Developer Relations

If you missed the April 30 webinar with Kyle Kingsbury, the creator of Jepsen and Karthik Ranganathan, CTO at YugaByte, you can check out the playback here and download the slides here.

Kyle did a great deep-dive into his testing approach, research methodology and what he found when applied to YugabyteDB. If you are looking for an unvarnished and honest assessment of YugabyteDB’s consistency and transactional claims, this video is a must watch!

To recap, YugabyteDB 1.2 was Jepsen certified last month. You can read the official report here, plus YugabyteDB’s take on the process and the results here. A DIY version of the Jepsen test suite has been integrated into our CD/CI pipeline since 2018.

Questions and Answers

Here’s a few of the questions that came up in the Q&A portion of the webinar:

What’s your heuristic for assessing Jepsen tests that database vendors come up with internally? What is the most common spot that internal teams miss, maybe because they know it too well?

*  *  *

When a database is CP (Consistent and Partition Tolerant) and makes no claims in regards to availability, how do you detect that a cluster stopped working to the extent that it should fail the test?

*  *  *

Do you stay strictly blackbox in your testing, that is, only reason about high-level invariance or will you sometimes extend Jepsen to peek at metrics or end points that the databases exposes and assert on them?

*  *  *

When do you expect users to be able to try out the new generators implementation?

Want the answers? Check out the playback!

What’s Next

  • Compare YugabyteDB in depth to databases like CockroachDB, Google Cloud Spanner and MongoDB.
  • Get started with YugabyteDB on macOS, Linux, Docker, and Kubernetes.
  • Contact us to learn more about licensing, pricing or to schedule a technical overview.

Related Posts

Jimmy Guerrero

VP Developer Relations

Related Posts

Get started in any cloud, container or data center