Making containers(programs) to run like water.
Making checkpointing facility available with any container (docker is used here to demonstrate)
This contains the required configuration that one would need to setup their system in order to make fault tolerant or have the facility of live migration with them.
Checkpoint restart a facility available in the docker only works with specific versions of docker. Rest it does not work and checking various debian system it does not work with the latest versions.
This configuration is a solution to all those problem and also a demo how can we do load balancing on server-side w.r.t cpu utilization
The below listed folders consists of the two implemented systems:
- simple_loop: It contains code for automated checkpoint and restore of a process on same node(container)
- automated_checkpoint_restore: It contains code to build a prototype fault tolerant system using automated checkpoint and restore mechanism.