Hi Rob,

Try to test the same results using bpbkar.b4patch446 binary and see what you
get.

maybe you want to try the same with the tar.b4patch446 as well.

Ayaz

#!/a.k.a/jazz
_____________________________________________________________
Ayaz Mudarris, Senior Consultant
Collective                      http://www.collectivetech.com
Tel  (312) 781-6200             eFax (253) 541-2141

> -----Original Message-----
> From: veritas-bu-admin AT mailman.eng.auburn DOT edu
> [mailto:veritas-bu-admin AT mailman.eng.auburn DOT edu]On Behalf Of Rob Worman
> Sent: Wednesday, April 04, 2001 10:27 AM
> To: 'veritas-bu AT mailman.eng.auburn DOT edu'
> Subject: [Veritas-bu] HP-UX 11.0 and NBU 3.2
>
>
> Hey All-
>
> We're running NBU 3.2 patch 466 on HP-UX 11.0 and we're seeing backup
> performance that seems unduly slow.
>
> NBU performance is, of course, a complex concept with lots of pieces that
> could slow it down, but it looks pretty simply when I see the following
> results of local writes to /dev/null:
>
>     #tar cf - /var > /dev/null
> (writes at 18GB/hr)
>
>     #/usr/openv/netbackup/bin/bpbkar -dt 0 -r 888 -nocont /var > /dev/null
> (writes at 8 GB/hr)
>
> We have noticed that disabling JFS Asynchronous I/O (via "echo 1 >
> /usr/openv/netbackup/AIO_READS_MAX") speeds up the above bpbkar
> to 11GB/hr.
>
> And for what it's worth,  I don't see an NBU slowdown when
> reading the raw
> device:
>
>    #dd if=/dev/vg00/rlvol8 of=/dev/null bs=32768
> (writes at 84 GB/hr)
>
>    #/usr/openv/netbackup/bin/bpbkar -dt 0 -r 888 -nocont
> /dev/vg00/rlvol8 >
> /dev/null
> (also writes at 84 GB/hr)
>
>
> Has anyone else already trod this path?
> TiA
> rob
>
> _______________________________________________
> Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
> http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
>