Author: "Rushforth, Tim" <TRushfor AT CITY.WINNIPEG.MB DOT CA>
Date: Fri, 1 Nov 2002 09:45:12 -0600
The number of objects backed up (11,854) does not match the # of objects restored (2687+1122) so you are not comparing the same thing. Does the backup include NDS? To do an accurate comparison, only
The large number of files are from the SYS volume which we don't restore on a daily basis. We'll try to test just the VOL1 and VOL2 backup though. We commented out the resourceutilization parameter i
Hi, We just installed Gigabit fiber NICs and an isolated gigabit fiber switch. Our restores have increased dramatically from 31.81 GB/hr (8.84 MB/s) to 56 GB/hr (15.6 MB/s). The backups are still aro
Author: "Seay, Paul" <seay_pd AT NAPTHEON DOT COM>
Date: Wed, 30 Oct 2002 00:02:14 -0500
Actually, the 2048 TCPWINDOWSIZE is not supported in NT to my knowledge. It is supported in W2K at SP1 or 2, cannot remember, with a registry hack. Someone else will have to give the particulars on t
Hi, Our backups are set for an absolute serialization. We backup and restore the same number of files and the same amount of data 42 GB. The TSM server is configured in a RAID 1E0 with a ServeRAID 4M
Author: Steve Schaub <Steve.Schaub AT HAWORTH DOT COM>
Date: Thu, 31 Oct 2002 05:52:31 -0500
Paul (or anyone who knows), If Vin has the logging mode set to rollforward, does that impact performance? I only suggest this because I have had a nagging slowdown in some operations and the only cha
Author: Rainer Wolf <rainer.wolf AT RZ.UNI-ULM DOT DE>
Date: Thu, 31 Oct 2002 13:52:25 +0100
Hello Vin, our tsm-server is just on gigabit and some of the clients are also there. My guess is that your restore - performance is just showing what your drive can read ( maybe magstar - drives ? ).
Hi, We are backing up to the TSM's 150 GB Diskpool and restoring from the 150 GB Diskpool as well. Although we haven't tried to back up straight to tape (the DLT library has 2 DLT IV drives), we stil
We've tried using 524MB and 131MB for our bufferpool size with no effect. Our logpool size is 8MB. We have 99 percent db hit percentage. We aren't using TSM mirroring on the db or log. The primaries
Author: "Rushforth, Tim" <TRushfor AT CITY.WINNIPEG.MB DOT CA>
Date: Thu, 31 Oct 2002 09:53:58 -0600
It's been mentioned that it is probably the RAID5 on your TSM storage pool. To rule out TSM, simply do a FTP from the client to the same disks on the server and see what throughput you get. Hi, We ju
Author: Nicholas Cassimatis <nickpc AT US.IBM DOT COM>
Date: Thu, 31 Oct 2002 12:13:14 -0500
Is the performance comparison from the stats ITSM gives at the end of the session? If so, are you giving us the network or the aggregate transfer? I'd expect the network transfer rate to be about the
Author: "Seay, Paul" <seay_pd AT NAPTHEON DOT COM>
Date: Wed, 30 Oct 2002 22:36:31 -0500
You have lots of RAM, but what is your bufferpool size. For this size of machine memory, I would use at least 256MB for the bufferpool size. Also beef up the logpool size as well. Your TSM DB and log
We used the bytes transfered and elapsed time to calculate the throughput. Here are the actual stats though: Backup: 10/25/2002 16:03:12 Total number of objects inspected: 11,858 10/25/2002 16:03:12
We are not using a RAID5 on ANY drives. They are all RAID1 or RAID1E0 (spanned RAID1E). Unfortunately I can't load an FTP server on the Netware server without a LOT of red tape. I'm not necessarily t