Summary

Application checkpoint/restart allows us to do live migration of LXC containers, greatly increasing their usability as an alternative cluster node type.

Release Note

Checkpoint, restart, and live migration of applications and containers is now supported.

Rationale

A great deal of support for application c/r is already implemented in out-of-upstream patchsets, and available at linux-cr.org. The purpose of this blueprint is simply to publish a PPA with the kernel and userspace packages needed for users to trivially test c/r, be it on their host, in a VM, or on EC2.

User stories

Mike is a user with a running mysql server. He wants to temporarily migrate the mysql instance to another machine while he upgrades the server's operating system.

John wants to run a cluster using containers. There are two long running jobs in LXC containers on the same physical machine. He migrates one container to another machine to increase performance for both jobs.

Assumptions

Design

Implementation

Test/Demo Plan

Unresolved issues

BoF agenda and discussion

UDS Natty discussion

== Application checkpoint/restart ==

Application checkpoint/restart in linux (linux-cr.org) provides the ability to checkpoint, restart, and migrate application and system containers. This provides a very lightweight mechanism for load-balancing in the cloud.

Actions:
 * Create a ppa with the kernel and userspace packages needed to experiment with c/r
   * Create a project in lp
   * When ppa is up, Gustavo will blog about how to use it


CategorySpec

CloudServerNAppCheckpointRestart (last edited 2010-11-09 17:01:35 by serge-hallyn)