Checkpoint/Restore track

Thursday, August 20, 2015 from 1:304:00pm
Metropolitan B
 

Checkpoint/restart technology is the basis for live migration as well as its traditional use to take a snapshot of a long-running job. This microconference will focus on the C/R project called CRIU and will bring together people from Canonical, CloudLinux, Georgia Institute of Technology, Google, Parallels, and Qualcomm to discuss CRIU integration with the various containers projects, its use on Android, performance and testing issues and, of course, to show some live demoes. See the Checkpoint/Restart wiki for more information.

Microconference Leader

Pavel Emelyanov

Proposals for this track

* Checkpoint and Restore of processes within kernel security mechanisms

A discussion about what to do for c/r of LSMs, selinux, and user namespaces.
Checkpoint/Restore 06/16/2015
Tycho Andersen

* CRIU in HPC

Why CRIU was used as the basis for checkpointing/restarting and migrating in a HPC research project. (slides)
Checkpoint/Restore 05/08/2015
Adrian Reber

* CRIU on PowerPC

Discussion about the CRIU port on PowerPC (slides)
Checkpoint/Restore 04/28/2015
Laurent Dufour

* CRIU Support in Docker for Native Checkpoint and Restore

Discussion and live demonstration of how CRIU can be used by Docker to checkpoint and restore containers. (slides)
Checkpoint/Restore 04/07/2015
Saied Kazemi

* P.Haul: live migrating with CRIU

A quick talk about p.haul project that live migrates containers using CRIU (slides)
Checkpoint/Restore 06/25/2015
Pavel Emelyanov

* Rebootless Kernel Update and its verification

Seamlessly udpate the Linux kernel by checkpointing the running applications.
Checkpoint/Restore 05/09/2015
Sanidhya Kashyap

* Using CRIU for Computer Architecture and Software Studies

How checkpoints can speed up result generation, how CRIU impacts measurements of application performance, and how to mitigate those impacts. (slides)
Checkpoint/Restore 04/30/2015
Christopher Covington