On Wed, Jun 16, 2010 at 3:39 AM, Fco Javier Garcia <cor...@javido.com> wrote: > The main problem is not performance (for a home server is not a problem)... > but what really is a BIG PROBLEM is when you try to delete a snapshot a > little big... (try yourself...create a big random file with 90 Gb of data... > then
This is reportedly fixed in build past snv_134. I believe there was a single thread that reduced throughput dramatically. I was really excited to play with dedup and started using it around b131. Even with 8gb RAM and 30gb L2ARC, it took about a day to destroy some snapshots. The regular expiration by zfs-auto-snapshot would stall the system for a few hours. Writes to dedup volumes were painfully slow, around 10k/s. I suspect that my DDT was larger than my L2ARC - I had a lot of data with dedup enabled. I've since done a send to another system and back to re-dup everything, which has restored performance at a cost of twice the space. -B -- Brandon High : bh...@freaks.com _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
