There was no better way to kick off KubeCon EU than with Data on Kubernetes Community Day on Monday, May 3. K8ssandra featured in a number of sessions presented by community members. Here we highlight some of these sessions, ranging from high-level visions for the future to detailed deep dives.
DBA to SRE
I had the chance (Patrick McFadin) to break the ice with my keynote, Going from DBA to SRE: Time to make the move. The role and skill of Database Administrator has been shrinking as we move from highly specialized to generalized IT in a cloud native world. This is a call to action for DBAs to embrace the future and transition to the incredibly important skill of site reliability engineering. You have the skills and background to make a huge impact! There are new skills to learn but there are great resources available.
Data Services with Stargate
Later on, Jeff Carpenter talked about the Stargate API Gateway and why we’ve included it as part of the K8ssandra project: to make Cassandra more accessible to developers. In Data Services for the Masses, Jeff shared how Stargate nodes can be scaled alongside a Cassandra cluster to help offset query loads. This gives you more control over the scaling of compute vs. storage in K8ssandra-provisioned clusters. As with many innovations in computer science, the ideas behind Stargate are not completely new, drawing on previous work done by Cassandra contributors from companies such as Instagram and SimpleReach.
Cassandra on Kubernetes
Christopher Bradford took the stage next, digging into how two disparate distributed systems, Apache Cassandra and Kubernetes, can play to each other’s strengths to build a next-generation data platform. In his talk, Finding peaceful co-existence between Cassandra and Kubernetes, Chris explored the history, trials, and tribulations of bringing Cassandra to Kubernetes. The result has been a better user experience and simpler management surface. This also included some aspirational moments exploring what the future holds for K8ssandra and multi-cluster deployments.
Modernizing Databases for Kubernetes
Rick Vasquez shifted our perspectives with my favorite talk of the day. He challenged databases to go all-in on Kubernetes by embracing the cloud-native ecosystem in his talk, A Call for DBMS to Modernize on K8’s. Rick notes that operators are a move in the right direction, but ultimately not enough. Databases built for Kubernetes should integrate natively with compatible storage backends and network layers. Every database built for Kubernetes needs to embrace scaling, elasticity, and self-healing.
K8ssandra and OpenEBS
Rounding things out, Eric Zietlow and Aleks Volochnev provided a great demonstration in their talk, Deploying open cloud-native data using K8ssandra. They showed how to deploy K8ssandra alongside OpenEBS. This approach allows users to leverage the speed and flexibility of OpenEBS storage alongside the effortless deployment of K8ssandra. Storage is one of the first problems to solve when building stateful workloads and OpenEBS is striving to make that easy for cloud agnostic deployments.
I want to end with a huge shout out to the Data On Kubernetes Day organizers and community. It was a phenomenal event with amazing sessions and plenty of proof that Kubernetes is ready for stateful workloads. If you missed any sessions, the full set is available on YouTube.
We’re looking forward to the next event and contributing more to the projects that enable this group. Like every technical community, it thrives on sharing what has been learned. If you are interested in presenting or participating in community discussions, the DoK website has a lot of ways to get involved. See you there!