10ksim/deployment
Alberto Soutullo 8cff64566b
Copy file to docker image
2024-02-09 13:31:13 +01:00
..
gossipsub Update repo with gowaku + podmonitor 2024-02-01 11:18:55 +01:00
gowaku Changed dns resolve checking time async and clean structure. 2024-02-09 12:33:33 +01:00
nwaku Changed dns resolve checking time async and clean structure. 2024-02-09 12:33:33 +01:00
shared/publisher_image Copy file to docker image 2024-02-09 13:31:13 +01:00
README.md Update repo with gowaku + podmonitor 2024-02-01 11:18:55 +01:00
prepare_env.sh Added deployment scripts 2023-12-04 19:21:29 +01:00

README.md

Automated Deployment

While the Python tool is not ready, we are using a sort of bash automated deployment script.

Before running, make sure to prepare_env.sh all machines that will be used in the Kubernetes cluster.

In the gossipsub folder, you can find the deploy.sh script, which can be used either for the Nim, Go and Rust libp2p implementations that were use for the following results.

The deploy.sh script will call kubectl, so it is assumed that you have a Kubernetes cluster configured. The results were extracted using kube-prometheus-stack helm chart, and the get_logs.sh script is used to obtain the logs of all PODs that will be used to get the time plots.

Both nwaku and gowaku folder have the same, but with small variances to work with nWaku. Multi-container per pod is now a bit outdated, so please use 1 container per POD when asked in deploy.sh.