On Mon, Dec 22, 2014 at 02:04:37PM -0500, J. Bruce Fields wrote:
It'd also be nice to see any proposals for a completely correct
solution, even if it's something that will take a while. All I can
think of is protocol extensions, but that's just what I know.
I tried to think a little about
On 12/22/2014 06:41 PM, Jeff Darcy wrote:
An alternative would be to convert directories into regular files from
the brick point of view.
The benefits of this would be:
* d_off would be controlled by gluster, so all bricks would have the
same d_off and order. No need to use any d_off mapping
Please review http://review.gluster.org/9332/, as it undoes the
introduction of itransform on d_off in AFR. This does not solve
DHT-over-DHT or other future use cases, but at least fixes the regression
in 3.6.x.
Thanks
On Tue Dec 23 2014 at 10:34:41 AM Anand Avati av...@gluster.org wrote:
Using GFID does not work for d_off. The GFID represents and inode, and a
d_off represents a directory entry. Therefore using GFID as an alternative
to d_off breaks down when you have hardlinks for the same inode in a single
directory.
Good point. So what *can* we do locally on a brick to
The birthday paradox says that with a 44-bit hash we're more likely than
not to start seeing collisions somewhere around 2^22 directory entries.
That 16-million-entry-directory would have a lot of collisions.
This is really the key point. The risks of the bit-stealing approach
have been
On Mon, Dec 22, 2014 at 09:30:29AM -0500, Jeff Darcy wrote:
The birthday paradox says that with a 44-bit hash we're more likely than
not to start seeing collisions somewhere around 2^22 directory entries.
That 16-million-entry-directory would have a lot of collisions.
This is really the
An alternative would be to convert directories into regular files from
the brick point of view.
The benefits of this would be:
* d_off would be controlled by gluster, so all bricks would have the
same d_off and order. No need to use any d_off mapping or transformation.
I don't think a
On Mon, Dec 22, 2014 at 09:30:29AM -0500, Jeff Darcy wrote:
By contrast, the failure mode for the map-caching approach - a simple
failure in readdir - is relatively benign. Such failures are also
likely to be less common, even if we adopt the *unprecedented*
requirement that the cache be
On Mon, Dec 22, 2014 at 09:30:29AM -0500, Jeff Darcy wrote:
By contrast, the failure mode for the map-caching approach - a simple
failure in readdir - is relatively benign. Such failures are also
likely to be less common, even if we adopt the *unprecedented*
requirement that the cache be
On 12/17/2014 05:04 AM, Xavier Hernandez wrote:
Just to consider all possibilities...
Current architecture needs to create all directory structure on all
bricks, and has the big problem that each directory in each brick will
store the files in different order and with different d_off values.
Just to consider all possibilities...
Current architecture needs to create all directory structure on all
bricks, and has the big problem that each directory in each brick will
store the files in different order and with different d_off values.
This is a serious scalability issue and have
Sent: Tuesday, December 16, 2014 11:46:46 AM
Subject: Re: [Gluster-devel] Readdir d_off encoding
On 12/15/2014 09:06 PM, Anand Avati wrote:
Replies inline
On Mon Dec 15 2014 at 12:46:41 PM Shyam srang...@redhat.com
mailto:srang...@redhat.com wrote:
With the changes present in [1
On 12/16/2014 03:21 PM, J. Bruce Fields wrote:
On Tue, Dec 16, 2014 at 11:46:46AM -0500, Shyam wrote:
On 12/15/2014 09:06 PM, Anand Avati wrote:
Replies inline
On Mon Dec 15 2014 at 12:46:41 PM Shyam srang...@redhat.com
mailto:srang...@redhat.com wrote:
With the changes present in [1]
With the changes present in [1] and [2],
A short explanation of the change would be, we encode the subvol ID in
the d_off, losing 'n + 1' bits in case the high order n+1 bits of the
underlying xlator returned d_off is not free. (Best to read the commit
message for [1] :) )
Although not
14 matches
Mail list logo