Dynamic SLA Monitoring With Envoy

Install

This repository relies on Minikube and GCP to emulate microservice scenarios. We primarily use Ubuntu 18.04 for our tests. For convenience, dependencies can be installed with the ./tools/setup.sh command. It contains all the steps necessary to set up the testing environment.

Package Dependencies (Minikube)

• kubernetes to administrate and run web applications in a cluster
• minikube to run a local Kubernetes cluster
• docker as the container driver back end of Minikube
• bazel to build Wasm filters for Envoy
• istio to manage Envoy and its filters
• python3.6 to run the scripts

Package Dependencies (GCP)

TBD

Python Dependencies

• prometheus-api-client to query Prometheus

Running Experiments

Setup

Once everything is installed, the Kubernetes cluster can be started with ./kube_env.py --setup if you are running on minikube or ./kube_env.py --setup -p GCP if you are running on google cloud. If you are running on google cloud, add -p GCP to all subsequent commands as well. Starting the cluster will take a few minutes.

We provide a pre-built filter. It can be deployed with the ./kube_env.py --filter-name webassemblyhub.io/fruffy/test-filter --deploy-filter command. We are currently working on a solution that deploys locally-built filters.

Demo

Once the filter has been successfully installed, it is possible to run experiments with ./run_experiments.py --num-experiments 1.

Teardown

Remove the filter

./kube_env.py --filter-name webassemblyhub.io/fruffy/test-filter --undeploy-filter

Remove the bookinfo application

./kube_env.py --remove-bookinfo

Tear down the cluster

./kube_env.py --clean