[openstack-dev] [Ceph] Is it necessary to flatten Copy-on-write cloning for RBD-backed disks?

Clint Byrum clint at fewbar.com
Tue Jun 9 05:06:03 UTC 2015


Excerpts from Kun Feng's message of 2015-06-08 20:34:51 -0700:
> Hi all,
> 
> I'm using ceph as storage backend for Nova and Glance, and merged the
> rbd-ephemeral-clone patch into Nova. As VM disks are Copy-on-write clones
> of a image, I have some concerns about this:
> 
> 1. Since hundreds of vm disks based on one base file, is there any
> performance problems that IOs are loaded on this one paticular base file?
> 

Unless you have no ram available to VFS cache on the OSD's, this is
fine. Blocks will be evenly spread to each OSD, and since these are
likely to be super popular blocks, they'll likely all be served from
RAM most of the time.

> 2. Is it possible that the data of base file is damaged or PG/OSD
> containing data of this base file out of service, resulting as all the VMs
> based on that base file malfunctioned?

You probably will want to read this and see if that answers your
question:

http://ceph.com/docs/master/architecture/#data-consistency



More information about the OpenStack-dev mailing list