okay , i forgot ;fs flush ; fs flushvolume  in my example

i see with iostat an average write speed of 32-38 MB/sec on the fileserver , so why should the speed be lower ? but i will try the afsfsperf tool to collect the data .

can you post your config changes on that System ? BosConfig , tcp changes ... just to test with the same Settings .

the afs Version would also be good .

Sven

-------------------------------------------------------------------------------------------------------------------------
Dept. A141,  TG/SSG EMEA AIS
Development Leader Stonehenge
IBM intranet ---> http://w3.ais.mainz.de.ibm.com/stonehenge/
internet ---> http://www-5.ibm.com/services/de/its/filestore.html
Phone (+49)-6131-84-3151
Fax      (+49)-6131-84-6708
Mobil   (+49)-171-970-6664
E-Mail : [EMAIL PROTECTED]



Jimmy Engelbrecht <[EMAIL PROTECTED]>
Sent by: [EMAIL PROTECTED]

21.01.2004 11:44

To
Sven Oehme/Germany/[EMAIL PROTECTED]
cc
Aaron Stanley <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
Subject
Re: [OpenAFS] High Speed AFS Access





"Sven Oehme" <[EMAIL PROTECTED]> writes:

> for example  time dd if=/dev/zero of=/afs/test/testfile.dd count=50 bs=1M
> ; sync
>
> and then used the calculator  :-)

I am sorry to disappoint you but thats completly useless since you this way
only test performance of the local cache.

try afsfsperf from the arla-package for usefull numbers.

/Jimmy

Reply via email to