edgerun-galileo-faas 0.0.1.dev9

Galileo FaaS

Galileo FaaS

This project aims to provide easy to use functions to monitor a running K3S cluster, including telemetry data and traces. The provided functions can help to build intelligent runtime optimizations, i.e., load balancing or scheduling strategies.

For this environment to work, you need to have the following setup:

• a running K3S cluster
• an etcd instance
• redis for storing nodeinfos and as pub/sub framework
• telemd daemons deployed to monitor resource usage
• telemd-kubernetes-adapter daemon deployed in the Kubernetes cluster that monitors all Pods an emits events upon changes.

In case you want to start a full-fledged experiment with galileo, you will also need:

• InfluxDB to store telemetry, traces and events
• MariaDB for experiment metadata

Visit the galileo-experiments project to get further information on setting up an experiment environment.

Example

Look for an example on how to quickly start all services under galileofaas.cli.main.

Installation

Create a Python 3.9 virtual environment and install requirements.txt and requirements-dev.txt (see Makefile).

Hint

Currently you have to install the following dependencies from source: edgerun/faas and edgerun/skippy-core (model-updates branch).

Environment Variables - Galileo FaaS

Variable	Default	Description
galileo_faas_logging	DEBUG	Log level (DEBUG, INFO, WARN, ERROR)
galileo_faas_telemetry_window_size	60	the time window that will be cached, in s
galileo_faas_trace_window_size	60	the time window that will be cached, in s
galileo_faas_redis_host	localhost
galileo_faas_redis_port	6379
galileo_faas_redis_password	optional	Password for Redis
galileo_faas_influxdb_url	http://localhost:8086	URL to InfluxDB v2 instance
galileo_faas_influxdb_token	token	Token for InluxDB instance
galileo_faas_influxdb_org	org	InfluxDB organziation to save experiments in
galileo_faas_influxdb_timeout	10_000	Timeout for InfluxDB connections (in ms)
galileo_faas_k8s_config	local	Configures where the galileo-faas program is started (local or incluster)
galileo_faas_etcd_host	localhost	Host of the Etcd instance (relevant for load balancer to get weights)
galileo_faas_etcd_port	2379	Port of the Etcd instance
galileo_faas_async_pod	False	Set to True if Pod creation should by async