Thank you for sharing the video about running postgres on K8s. Very useful and contains good pointers, especially not running the database on the same host as the application, not provisioning persistent volumes locally, and where possible, run PG on it's own node and allocate all the resources needed and efficient configurations for storage, e.g., SSD/NVme.
@@simplyblock Yeah, will do. Running PG on K8s was something I had wondered in my use case, admittedly a small app, the application will fail if the DB goes offline. So I can see it's good to take advantage of features K8s provide, e.g., HA, etc. However, getting the right settings for PG on K8s is definitely not for the faint of heart, and you alluded to that in the talk and provided links for further research and talking to expert consultants as appropriate.
@@jeromeeusebiusstarting on a small app is totally fine. You can start co-located in the same node pool and move it to a separate one as the database grows. That's the beauty of orchestration. Small change and it happens automagically.
Thank you for sharing the video about running postgres on K8s. Very useful and contains good pointers, especially not running the database on the same host as the application, not provisioning persistent volumes locally, and where possible, run PG on it's own node and allocate all the resources needed and efficient configurations for storage, e.g., SSD/NVme.
Glad you found it helpful 🔥 Keep an eye on upcoming webinars 😊
@@simplyblock Yeah, will do. Running PG on K8s was something I had wondered in my use case, admittedly a small app, the application will fail if the DB goes offline. So I can see it's good to take advantage of features K8s provide, e.g., HA, etc. However, getting the right settings for PG on K8s is definitely not for the faint of heart, and you alluded to that in the talk and provided links for further research and talking to expert consultants as appropriate.
@@jeromeeusebiusstarting on a small app is totally fine. You can start co-located in the same node pool and move it to a separate one as the database grows. That's the beauty of orchestration. Small change and it happens automagically.