Skip to content

Commit

Permalink
add base scripts to test GKE elasticity (#176)
Browse files Browse the repository at this point in the history
we will do more work with this to integrate the flux operator
in too.

Signed-off-by: vsoch <vsoch@users.noreply.github.com>
Co-authored-by: vsoch <vsoch@users.noreply.github.com>
  • Loading branch information
vsoch and vsoch committed May 23, 2023
1 parent 69dae39 commit e54ac7b
Show file tree
Hide file tree
Showing 3 changed files with 607 additions and 0 deletions.
31 changes: 31 additions & 0 deletions examples/elasticity/google/autoscale/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
## Google GKE Autoscaling

> Using Google Cloud APIs to autoscale + Flux Operator autoscaling
These classes and assets are intended to help with autoscaling experiments.

Tests of elasticity, for GKE and the Flux Operator, will help us move us a tiny step toward the goal of
trying to control both the GKE cluster scaling and MiniCluster scaling from the same
script. This way, theoretically, some other algorithm (that is smarter than my manual calls) can do it too.
See the [elasticity page](https://flux-framework.org/flux-operator/tutorials/elasticity.html) for early work.

To start, however, we just want to look at some basic elasticity or scaling for Kubernetes by itself.
All experiments use the shared class in [fluxcluster.py](fluxcluster.py), and an example experiment
is [test-scale.py](test-scale.py).


### Dependencies

We will need the google cloud APIs

```bash
$ python -m venv env
$ source env/bin/activate
# https://github.com/googleapis/python-container
$ pip install google-cloud-container
$ pip install kubernetes
```

### Experiments

The scripts are provided here to do experiments, and you can see results [here](https://github.com/converged-computing/operator-experiments/tree/main/google/autoscale/run1).

0 comments on commit e54ac7b

Please sign in to comment.