docs: Document running at massive scale. Fixes #6443 #6459
Conversation
Signed-off-by: Alex Collins <alex_collins@intuit.com>
Codecov Report
@@ Coverage Diff @@
## master #6459 +/- ##
==========================================
- Coverage 48.39% 48.32% -0.07%
==========================================
Files 259 259
Lines 18675 18675
==========================================
- Hits 9037 9025 -12
- Misses 8647 8662 +15
+ Partials 991 988 -3
Continue to review full report at Codecov.
|
docs/running-at-massive-scale.md
Outdated
| * Use the Emissary executor (>= v3.1). This does not make any Kubernetes API requests. | ||
| * Limit the number of concurrent workflows using parallelism. | ||
| * Rate-limit pod creation [configuration](workflow-controller-configmap.yaml) (>= v3.1). | ||
| * Set `DEFAULT_REQUEUE_TIME=1m`. |
There was a problem hiding this comment.
Link to https://github.com/argoproj/argo-workflows/blob/master/docs/environment-variables.md for how to set this?
docs/running-at-massive-scale.md
Outdated
|
|
||
| Where Argo has a lot of work to do, the Kubernetes API can be overwhelmed. There are several strategies to reduce this: | ||
|
|
||
| * Use the Emissary executor (>= v3.1). This does not make any Kubernetes API requests. |
There was a problem hiding this comment.
Unless the templates are k8s custom resources
| @@ -0,0 +1,40 @@ | |||
| # Running At Massive Scale | |||
|
|
|||
| Argo Workflows is capable of running 1,000s of workflows a day, each with 10,000s of nodes. But you'll need to do some | |||
There was a problem hiding this comment.
Link to the survey or stress test suite?
Co-authored-by: Michael Crenshaw <michael@crenshaw.dev>
Co-authored-by: Yuan Tang <terrytangyuan@gmail.com>
Signed-off-by: Alex Collins alex_collins@intuit.com
Fixes #6443