[sheepdog] Users inputs for new reclaim algorithm, please

Liu Yuan namei.unix at gmail.com
Tue Mar 18 06:57:48 CET 2014


On Tue, Mar 18, 2014 at 07:19:49AM +0900, MORITA Kazutaka wrote:
> At Tue, 18 Mar 2014 00:02:18 +0800,
> Liu Yuan wrote:
> > 
> > Keep old algorithm is a bottom line to me. Either sheep option or static #ifdef
> > looks fine to me. We can later refine it to be dynamically pluggable.
> > 
> > New algorithm looks to me more a partial solution than a generic one, compared
> > with old algorithm. So keep old algorithm really make sense especially to those
> > who seldom delete snapshots.
> 
> Why partial?  The updated snapshot-object-reclaim branch fixes the
> problem you mentioned, and the new algorithm frees more spaces than
> the old one even with your example scenario.
> 

The new version looks much better than last one and performance drop is around
10% for first write and I didn't experence space increase for clone of deletion.

So if Valerio won't find any problem, it is fine to me to merge it.

Thanks
Yuan



More information about the sheepdog mailing list