Actions
Bug #21167
opencontainer checkpoint/restore
Story points:
-
Description
It would be interesting to be able to checkpoint a Crunch container, then restore it later. A couple cases where this would be useful:
- avoid losing progress when a spot instance is reclaimed by checkpoint periodically and/or when the "about to be reclaimed" notification happens (depends on how fast you can save the process)
- Start a process on an medium sized instance type, and automatically checkpoint and restore to a larger/smaller instance with more/less RAM if the RAM usage goes above/below a threshold.
https://memverge.com/ (commercial)
https://criu.org/Main_Page (open source)
"docker checkpoint" is an experimental feature that uses CRIU, https://docs.docker.com/engine/reference/commandline/checkpoint/
Looks like it's still classified as experimental (requires enabling "experimental" features in the docker config file) but is present in current docker releases.
Some more information here: https://criu.org/Docker
Updated by Peter Amstutz about 1 year ago
- Status changed from New to In Progress
Updated by Peter Amstutz about 1 year ago
- Description updated (diff)
- Subject changed from Process checkpoint/restore to container checkpoint/restore
Updated by Peter Amstutz about 1 year ago
- Category set to Crunch
- Description updated (diff)
Actions