Using 3592/jaguar tapedrives with tsm 5.2 and 5.3.
LANFree backup of a SAP database (1.8TB) using the TDP for R3 writing to
1 tapedrive.
2 HBA's for disk and 2 HBA's for tape (2gbit). Tape and disk go via
different fabrics.
BKI1215I: Average transmission rate was 241.800 GB/h (68.779 MB/sec).
BKI1227I: Average compression factor was 1.00.
BKI0020I: End of program at: Tue Sep 12 07:58:53 DFT 2006 .
BKI0021I: Elapsed time: 07 h 22 min 22 sec .
BKI0024I: Return code is: 0.
When writing to 2 3592 tapedrives we get a throughput of ~120MB/sec in
total.
Jeroen
-----Original Message-----
From: ADSM: Dist Stor Manager [mailto:ADSM-L AT VM.MARIST DOT EDU] On Behalf Of
goc
Sent: Friday, October 27, 2006 10:31 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: [ADSM-L] LANFree report !
to continue (i know i'm boring you to death with this) i used
RESOURCETUIL 10 and MAXNUMMP 3 and started the whole process again, the
gain was 6.4 hours less time to backup 3.4TB
Total number of objects inspected: 250
Total number of objects backed up: 250
Total number of objects updated: 0
Total number of objects rebound: 0
Total number of objects deleted: 0
Total number of objects expired: 0
Total number of objects failed: 0
Total number of bytes transferred: 3.39 TB
LanFree data bytes: 3.39 TB
Data transfer time: 4,661.02 sec
Network data transfer rate: 781,906.64 KB/sec
Aggregate data transfer rate: 79,216.06 KB/sec
Objects compressed by: 0%
Elapsed processing time: 12:46:46
i would like to hear some ideas and thoughts from you girls and guys :-)
thanks
----- Original Message -----
From: "goc" <goran.k AT vip DOT hr>
To: <ADSM-L AT VM.MARIST DOT EDU>
Sent: Thursday, October 26, 2006 12:21 PM
Subject: LANFree report !
> okay, i've managed to setup something and here is the result of the
> test (LOL yea, testing with 3.4TB oracle statistical database) ...
> anyways
>
>
> 10/26/06 10:59:29 --- SCHEDULEREC STATUS BEGIN
> 10/26/06 10:59:29 Total number of objects inspected: 249
> 10/26/06 10:59:29 Total number of objects backed up: 211
> 10/26/06 10:59:29 Total number of objects updated: 0
> 10/26/06 10:59:29 Total number of objects rebound: 0
> 10/26/06 10:59:29 Total number of objects deleted: 0
> 10/26/06 10:59:29 Total number of objects expired: 0
> 10/26/06 10:59:29 Total number of objects failed: 0
> 10/26/06 10:59:29 Total number of bytes transferred: 3.39 TB
> 10/26/06 10:59:29 LanFree data bytes: 3.39 TB
> 10/26/06 10:59:29 Data transfer time: 3,197.04 sec
> 10/26/06 10:59:29 Network data transfer rate: 1,139,038.36
KB/sec
> 10/26/06 10:59:29 Aggregate data transfer rate: 52,360.73
KB/sec
> 10/26/06 10:59:29 Objects compressed by: 0%
> 10/26/06 10:59:29 Elapsed processing time: 19:19:07
>
> client 5.3.4.0 and storage agent on HPUX 11 - HP RP8400 8proc/8GB RAM
> (2 1Gb FC adapters) server AIX 5.3.4 with LTO3 drives in 3584, the
> files are on HP XP12000 storage box connected to HP SAN COREEDGE 4Gb
> switches
>
> i was thinking in RESOURCEUTILization and MAXNUMMP ... to speed up the
> whole process maybe since i did not touch the above default values ...
> or simply to upgrade FC adapters to 2 or 4Gb :-), no compression
> included
>
> the process included continues access volumes, splitting on another
> XP12000 storage and then backing up from splitted volumes, the split
> od 3.4TB lasted for 33 minutes
>
> (Executing scheduled command now.
> 10/25/06 15:07:58
> Executing Operating System command or script:
> /ca/START.sh
> 10/25/06 15:40:22 Finished command. Return code is: 0
> 10/25/06 15:40:22 --- SCHEDULEREC OBJECT BEGIN SDB1_BC_BCKP_SCHED01
> 10/25/06 15:00:00)
>
> ------------------
> goran
> VIPNET
>
>
disclaimer.txt
Description: Text document
|