Gavin, Here is a quick summary.
This is a single machine it is running file, storage and director daemon. The postgres database is on a separate machine. All services are communicating via IP not the loopback. There does not seem to be any difference in speed if I use loopback or IP. This is a dual quad core xeon with 4Gig of mem We have 8 SATA drives connected to a 3ware 9690SA-8I raid controller in a Raid-6 configuration. Card has 512M RAM and a battery backup unit. The issue I am running into is during a full backup I am getting around 9.5Meg/s vs. the 85Meg/s on a restore. You would think the raid array would be able to read faster than it can write. -H -----Original Message----- From: Gavin McCullagh [mailto:gavin.mccull...@gcd.ie] Sent: Wednesday, December 09, 2009 11:38 AM To: bacula-users@lists.sourceforge.net Subject: Re: [Bacula-users] Network transfer Speed On Wed, 09 Dec 2009, Hayden Katzenellenbogen wrote: > I started an 800G restore. Which is half of my 1.7TB backup. > > I ran iostat while it was restoring and was getting between 50 and 140 > Meg/s. I would say the average is around 85Meg/s. Is that 85MByte/sec across the network or from local backup to local disk? Do you mind if I ask what media you're using at each end (tapes or disks, sata/scsi/lto-X/....). That sounds pretty quick to me. Gavin ------------------------------------------------------------------------ ------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users ------------------------------------------------------------------------------ Return on Information: Google Enterprise Search pays you back Get the facts. http://p.sf.net/sfu/google-dev2dev _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
