Hi Jummo

Thanks for your answer, comments inline...

Am 05.07.2012 10:08, schrieb Jummo:
>
> On Mon, 2 Jul 2012, Mario Moder wrote:
>> What I don't understand is the result of the next two tests (with bacula
>> block structure):
>>
>> - Fastest zero test (4.294 GB): 84.21 MB/s
>> - Fastest random data test (4.294 GB): 81.03 MB/s
>>
>
> We got higher rates with btape test, zero 12 GByte with 121 MByte/s and
> random 12 GBytes with 106.4 MByte/s, but with a real backup job we don't
> hit more than 85 MByte/s. We use dataspool to local 15k SAS drives to
> feed the tape drive. Do you have real backup job throughput rates?

Despooling elapsed time = 00:09:21, Transfer rate = 78.77 M Bytes/second

That's what I just got from a 44.14 GB Backup from a local disk tape 
spool. It was a good-compressable SQL Export file, should at least be 
2:1-compressable by hardware because it is 10:1-compressable by software.

>
> btape output: http://pastebin.ca/2167291

Yeah, your btape results look better than mine, so I guess there's some 
hardware bottleneck here.

Do you have any non-default values set for your tape drive (block sizes, 
max file sizes, etc.) which could have an influence on the btape test?

Interesting though that you also get no more than 85 MB/s on real 
backups. Are your 85 MB/s the despooling rate or the total backup rate?


Greetings

Mario


------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Bacula-users mailing list
Bacula-users AT lists.sourceforge DOT net
https://lists.sourceforge.net/lists/listinfo/bacula-users