PRESS PLAY
The Data on Kubernetes Community (DoKC) is where users go to run data on Kubernetes. We facilitate the creation and sharing of best practices to help users advance in their DoK journey. Here you can enjoy the audio from our livestreams and meetups. Learn more at https://dok.community/
A Kubernetes service can expose your database; your next deployment review just got more interesting.
Press play if you want
Short reviews from the PlayNext desk, based on the episodes we processed.
A candid government case study on database performance, access control, and practical Kubernetes tradeoffs.
Sebastian's presentation looks at the practical work of running databases on Kubernetes inside the Dutch government, opening with the "break the glass" problem and then moving into database performance on on-premise Ceph. Its strength is the level of operational detail.
A three-person data team turns a Glue-to-Argo migration into a grounded orchestration case study.
Pay It Gov's move from AWS Glue to Argo Workflows and Hera gives this Data on Kubernetes episode a clear practical spine: a data team working to bring pipeline orchestration back under control after a legacy setup and an earlier warehouse iteration proved difficult to manage.
Specialists walking through software's tradeoffs, one system at a time.
Founders and engineers on AI coding agents like Claude Code, and supply chain security.
The engineering decisions behind code, agents, and control.
Python tools, practical trade-offs, and the occasional programming joke.
Matt Stratton interviews DevOps practitioners on security, platform engineering, and trusting AI-written code.
["Sebastian Mannem says the Dutch government tested Cloud Native PG on OpenShift with Ceph and saw about 1,300 TPS at 62 milliseconds of latency.", "He argues that a \"break the glass\" path based on client certificates and mTLS fits app-to-database access better than LDAP-heavy human logins.", "He then shows that separating write-ahead log storage, adding CPU and RAM, and even turning fsync off changed the benchmark far more than simple resource increases, with fsync-off doubling TPS and halving latency.", "He also compares Azure, Azure VMs, and Azure Postgres, describes his `pg_tps_optimized` tool, and ends with open-source recommendations for teams starting databases on Kubernetes."]
Matt Manzenski of PayIt Gov says the team moved data pipelines from AWS Glue to Argo Workflows and Hera after a year of Glue work in 2022 produced only two fact tables for BI users.
Robert Hodges explains how a Kubernetes service can accidentally create a public database endpoint.
The Data on Kubernetes Community town hall presents Spark on Kubernetes through a migration from YARN to Kubernetes and experience at Netflix.
Kafka began at LinkedIn in 2010, when roughly 70 million users and rapidly expanding features strained data infrastructure.
A stateful protocol is counterintuitive: TCP remembers the connection and requests dropped packets again, while UDP does not.
The Data on Kubernetes Community's first "Outer Nerd" live stream, after 155 prior streams, brings together Abhi Vaidyanatha, Fabian Met and Chase Christensen to explain why Dungeons & Dragons matters.
This episode discusses edge computing with K3S and ARM in the Data on Kubernetes ecosystem, featuring guest Sergio Mendez.
A panel discussion about running databases on Kubernetes, featuring experts sharing experiences and best practices.
A community livestream discussing best practices for running data on Kubernetes, with announcements about an upcoming conference and an interview with a software engineer.