Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[Bacula\-users\]\s+LTO3\s+performance\s*$/: 7 ]

Total 7 documents matching your query.

1. [Bacula-users] LTO3 performance (score: 1)
Author: "Sergio Belkin" <sebelk AT gmail DOT com>
Date: Fri, 2 Jan 2009 16:39:53 -0200
Hi, I have an IBM Type 8765-1UXD LTO3, is working but not at its best performance. As wikipedia its Max Speed (MB/s) should be Max Speed 80 MB/s and the actual is about 10 Mb/s. Could be some miscofi
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00006.html (12,307 bytes)

2. Re: [Bacula-users] LTO3 performance (score: 1)
Author: Jesper Krogh <jesper AT krogh DOT cc>
Date: Fri, 02 Jan 2009 20:14:42 +0100
Are you using spooling? Whats the speed measure using: dd if=/dev/zero of=/dev/nst0 bs=1M count=1000 -- Jesper -- _______________________________________________ Bacula-users mailing list Bacula-user
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00008.html (12,570 bytes)

3. Re: [Bacula-users] LTO3 performance (score: 1)
Author: "John Drescher" <drescherjm AT gmail DOT com>
Date: Fri, 2 Jan 2009 14:41:35 -0500
Is this over a network? If so is the network Gbit? Are you talking about an incremental backup or full? Is your database on the same disk as the source? Do you have spooling enabled? Compression? Yo
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00009.html (12,827 bytes)

4. Re: [Bacula-users] LTO3 performance (score: 1)
Author: "Jason A. Kates" <jason AT kates DOT org>
Date: Sat, 03 Jan 2009 03:57:45 -0500
For a speed test /dev/zero isn't the best item to use as the hardware compression will show how good it can be. I would test on files that aren't in the OS cache and will have representative level of
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00011.html (13,370 bytes)

5. Re: [Bacula-users] LTO3 performance (score: 1)
Author: Jesper Krogh <jesper AT krogh DOT cc>
Date: Sat, 03 Jan 2009 13:16:10 +0100
I agree, but if the test with /dev/zero ends with 10MB/s then there is definately something wrong in the system. So the speed test is indeed valid as a first shot test of the "drive only". Jesper --
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00014.html (12,777 bytes)

6. Re: [Bacula-users] LTO3 performance (score: 1)
Author: "Sergio Belkin" <sebelk AT gmail DOT com>
Date: Wed, 7 Jan 2009 10:55:39 -0200
The output test was: dd if=/dev/zero of=/dev/nst0 bs=1M count=1000 1000+0 records in 1000+0 records out 1048576000 bytes (1.0 GB) copied, 7.80926 seconds, 134 MB/s I have a server let say A that has
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00123.html (14,059 bytes)

7. Re: [Bacula-users] LTO3 performance (score: 1)
Author: Ralf Gross <Ralf-Lists AT ralfgross DOT de>
Date: Wed, 7 Jan 2009 20:57:02 +0100
Sergio Belkin schrieb: This looks ok. But sg0 is not the tape drive, it's a disk.. Show us the tapeinfo output of /dev/nst0 (while it's not in use) and the complete output (messages) of a bacula job
/usr/local/webapp/mharc-adsm.org/html/Bacula-users/2009-01/msg00134.html (13,865 bytes)


This search system is powered by Namazu