On Friday 12 October 2007 15:18, Alan Brown wrote: > On Fri, 12 Oct 2007, Kern Sibbald wrote: > > And I did as I said "I will increase this limit,". > > OK, I missed that bit/ > > > Also, as I said, I remain very skeptical about sizes greater than 500K, > > and there is even a certain amount of evidence from my own tests and from > > several other users that increasing the size above 128K makes no > > significant difference. > > FWIW the tape drives I'm using (HP LTO2 drives in a 60 slot robot) appear > to have 2Mb internal buffering. You may well be correct as long as the SD > can feed data to the drives fast enough to keep up, but even a SAN fabric > can become congested at times.
I suspect that tape drive chop up blocks to suit their needs so that they can index for doing quick forward spacing. From what I understand, older drives wrote in 512 byte chunks no matter what you did. This is unlikely to be true for modern drives, but then modern drives probably buffer up a meg or so of data before writing it, so I suspect that the block size is pretty irrelevant after a certain point. It all comes down to how fast Bacula can feed the data, and that is restricted either by the disk speed if you are spooling, or the comm line speed if you are writing directly. I'm starting to get contacts in hardware manufacturing companies, so will bring up this issue with them at some appropriate point, so we can get the point of view of a manufacturer. Regards, Kern ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users