Re: [CRIU] CRIU @Plumbers

[Saied Kazemi <saied-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>]

Since everyone is in agreement, let's call it checkpoint/restore then.
Restart often implies stop + start and, in fact, Docker already has a
restart command which does exactly that.

--Saied


On Tue, Feb 3, 2015 at 8:10 AM, Christopher Covington <cov-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
wrote:

> On 02/03/2015 09:58 AM, Pavel Emelyanov wrote:
> > On 02/02/2015 10:19 PM, Saied Kazemi wrote:
> >> Given the wide spectrum of CRIU use cases, I am also in favor of
> organizing a
> >> separate track for it.
> >>
> >> To that end, I can present and do a live demo of how CRIU was used to
> provide
> >> native Docker container checkpoint and restore.  This can be a full
> presentation
> >> by itself or combined with another presentation.
> >
> > That's great :) I'll start filling the page with data soon then. Feel
> free to
> > join and add the stuff you want.
> >
> >> A minor nit: for consistency, can we call it checkpoint/restore instead
> of checkpoint/restart?
> >
> > Google gives ~600k results for "checkpoint/restart" vs ~900k for
> "checkpoint/restore"
> > so I don't mind :)
>
> I prefer "restore" (perhaps a more migration focused term) myself, but I
> think
> there is some history behind "restart" (perhaps a more fault tolerance
> focused
> term) from at least the Berkeley Lab's Checkpoint/Restart project, which
> has
> software releases from 2005 and papers from 2002.
>
> http://crd.lbl.gov/departments/computer-science/CLaSS/research/BLCR/
>
> http://crd.lbl.gov/departments/computer-science/CLaSS/research/BLCR/checkpoint-restart-publications/
>
> Chris
>
> --
> Qualcomm Innovation Center, Inc.
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
> a Linux Foundation Collaborative Project
>