MiCADO is an auto-scaling framework for Docker applications. It supports autoscaling at two levels. At virtual machine (VM) level, a built-in Docker Swarm cluster is dynamically extended or reduced by adding/removing cloud virtual machines. At docker service level, the number of replicas implementing a Docker Service can be increased/decreased.
MiCADO requires a TOSCA based Application Description to be submitted containing three sections:
- The definition of the interconnected Docker services,
- The specification of the virtual machine and
- The implementation of scaling policy for both scaling levels.
To use MiCADO, first the MiCADO core services must be deployed on a virtual machine (called MiCADO Master) by an Ansible playbook. MiCADO Master contains Docker engine (configured as Swarm manager), Occopus (to scale VMs), Prometheus (for monitoring), Policy Keeper (to perform decision on scaling) and Submitter (to provide submission endpoint) microservices to realize the autoscaling control loops. During operation MiCADO workers (realised on new VMs) are instantiated on demand which deploy Prometheus Node Exporter, CAdvisor and Docker engine through contextualisation. The Docker engine of the newly instantiated MiCADO workers joins the Swarm manager on the MiCADO Master.