orchestration
This commit is contained in:
19
orchestration/TSYSK8S/TSYSK8S.md
Normal file
19
orchestration/TSYSK8S/TSYSK8S.md
Normal file
@@ -0,0 +1,19 @@
|
||||
# K8S at TSYS
|
||||
|
||||
## Introduction
|
||||
|
||||
In the near future (by end of 2024) we will need to spin our on premise heavy compute facility back online to run K8S.
|
||||
|
||||
This file is where we will document what that looks like. Mostly it will be used by R&D for the entire SDLC.
|
||||
|
||||
## Ecosystem
|
||||
|
||||
Probably
|
||||
|
||||
- https://github.com/spinnaker/spinnaker
|
||||
|
||||
as the overall orchestration layer.
|
||||
|
||||
Various K8S distributions have emerged.
|
||||
|
||||
We are (as of 10/12) a ways out from needing to worry about this in the critical path. We can go a long way with cloudron / cosmos / coolify and off the shelf docker containers (and even making some of our own containers for the tsys line of business application web sites).
|
9
orchestration/nonk8s.md
Normal file
9
orchestration/nonk8s.md
Normal file
@@ -0,0 +1,9 @@
|
||||
# Non k8s orchestration
|
||||
|
||||
## Introduction
|
||||
|
||||
Sometimes Subo and RR will want to run large workloads on bare metal. Not using k8s. We need to provide an option for that workload type. Probably managed via slurm.
|
||||
|
||||
## Options
|
||||
|
||||
- MAAS
|
Reference in New Issue
Block a user