[Gluster-users] Delay for geo-replication?

[Joe Landman]

On 05/27/2011 08:19 AM, Whit Blauvelt wrote:
> On Fri, May 27, 2011 at 12:12:31PM +0530, Kaushik BV wrote:
>
>>     The geo-replication crawls the volume continuosly for changes (does an
>>     intelligent crawl, i.e crawls down the fs hierarchy) only when there are
>>     changes beneath, and records those changes in the slave.
>
> Have you considered using the Linux kernel's INOTIFY function rather than
> crawling the directories? In theory it should be more efficient.

FAM did this stuff a decade ago for Irix and Unicos.  I think it was 
ported to Linux and then superceded by INOTIFY and related designs.

> On the other hand, by crawling the directories do you get the benefit of
> triggering any self-heal that may be needed automatically?

I wonder if it makes sense storing a local checksum and time stamp 
(governed by one master time sync unit in the gluster storage cluster), 
that would help determine what has changed.

-- 
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics, Inc.
email: landman at scalableinformatics.com
web  : http://scalableinformatics.com
        http://scalableinformatics.com/sicluster
phone: +1 734 786 8423 x121
fax  : +1 866 888 3112
cell : +1 734 612 4615