ADSM-L

Re: D2D vs. tape backups with TSM?

2004-08-02 09:15:03
Subject: Re: D2D vs. tape backups with TSM?
From: Otto Schakenbos <otto.schakenbos AT TELEFLEX DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Mon, 2 Aug 2004 15:14:44 +0200
Food for thought on multiple sessions and filebased backup:
If you got 2(or more) virt. tapes on one disk(array) and the client has
mounted them both and is pulling data from them this will probarly be
slower in throughput then when just reading from one virt. tape at a
time since sequantial reading is normally faster (esp. with ata) then
random.
In other words multiple mount points means the heads of your disk has to
move longer distances then when using a single mountpoint.
Thats why we have collocation turned on and only allow single sessions
on our file type backup pool.

maybe we are doing it all wrong anyway...

regards

Otto Schakenbos
System Administrator

TEL: +49-7151/502 8468
FAX: +49-7151/502 8489
MOBILE: +49-172/7102715
E-MAIL: oschakenbos AT teleflex DOT com

TFX IT-Service AG
Fronackerstrasse 33-35
71332 Waiblingen
GERMANY




TSM_User wrote:

I aggree about not using collocation but this customer has always collocated 
everything offsite and onsite.  Moving to ATA was step one.  Step two is 
getting them to turn off collocation on all the Large File data. Step three is 
going to be turnning off collocation for everything on disk.

"Rushforth, Tim" <TRushforth AT WINNIPEG DOT CA> wrote:We have always used 
compression going to Disk. We use exclude.compression
for things like .zip etc. You may want to use compressalways yes to avoid
resending data that grows.

We are not collocating at all - why would you want to? From a restore
perspective (using multi-session restore) it is better to have the data
spread out across multiple volumes.

-----Original Message-----
From: TSM_User [mailto:tsm_user AT YAHOO DOT COM]
Sent: July 28, 2004 8:29 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: D2D vs. tape backups with TSM?

We are using 25 GB volumes right now. We are also still collocating the
storage pools that use the file device class by node. This has worked out
fine for us. Sad to admit but I wasn't aware of the Technical Exchange
recommendation. Is there a white paper from that you could refer me to.

We are contemplating turning on node compression everywhere to also help
reduce disk space.

Also, I made mention in a previous post that we were reclaiming down to 50%
and that was fine. Well, like always when you make a comment like that it
makes you think and they you go look. I found that we were using around 16
TB's of ATA space in all when you look at the "In Use" numbers. When I
looked at the actual disk in use it was closer to 21 TB's of data. I am
currently reclaiming everything down to 40 and I plan to get down to 25
again. At that point I will compare the numbers and see how much I can
reduce the 21 TB's in use.

Also somewhat interesting information. We have found that the I/O
capabilities of the latest and greatest servers can really help push a lot
more data to disk. We have always been told by our disk vendor that the
bottleneck wasn't them. We ruled out many things except them. Finally we
looked at a more detailed performance monitor of our systems and we found
that the we were killing the processor during times when we were pushing a
lot of data to disk. With these new servers we see migrations from Fibre
disk to ATA disk at over 150 GB/hr. We do have 60 TB's of ATA space though
so we have a lot of disks to write to.


"Rushforth, Tim" wrote:
Just curious what size of file volumes are you using? We were originally
using 25 GB, and then I listened to the "Disk Only Backup Strategies"
Technical Exchange where they recommended 2-4 GB volumes.

Thanks,

Tim Rushforth
City of Winnipeg

-----Original Message-----
From: TSM_User [mailto:tsm_user AT YAHOO DOT COM]
Sent: July 27, 2004 6:41 PM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: D2D vs. tape backups with TSM?

Funny, we set ours down to 25% as well just to see what would happen. This
worked but we have since set all of the ATA Pools to 50% and we just leave
them there. Theoretically what could happen is we could be wasting twice as
much space but the fact is the volumes were going from 25% to 50% in a
matter of days and when we looked at how many volumes were between 25% and
50% in our environment we determined there was no need to reclaim down that
far. From all outward signs there was no issues with reclaiming down to 25%
we just didn't think it was worth doing the extra work to get back such a
small amount of disk. Disk is cheap, right! lol


"Rushforth, Tim" wrote:
We've set ours at 25%. We are just piloting an all disk backup pool for
some clients on one of our servers and for small files on another.


---------------------------------
Do you Yahoo!?
Y! Messenger - Communicate in real time. Download now.



---------------------------------
Do you Yahoo!?
New and Improved Yahoo! Mail - Send 10MB messages!





Attachment: otto.schakenbos.vcf
Description: Vcard

<Prev in Thread] Current Thread [Next in Thread>