Networker

Re: [Networker] Initial Report on Functionality improvements in N W7 using Adv_Fil e devices

2003-05-27 17:16:25
Subject: Re: [Networker] Initial Report on Functionality improvements in N W7 using Adv_Fil e devices
From: Bokkelkamp Ernst <ernst.bokkelkamp AT SIEMENS DOT COM>
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Date: Tue, 27 May 2003 19:26:47 +0200
This is not intended as an advert.

But check on www.netapp.com under NearStore R150.
AFAIK this system is currently limited to 24TB but available on (very)
special request with upto 96TB.
You may even be supprised about the cost and AFD was developed for this type
of system.

Bye
Ernie

-----Original Message-----
From: Joel Fisher [mailto:jfisher AT WFUBMC DOT EDU]
Sent: Tuesday, May 27, 2003 2:36 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] Initial Report on Functionality improvements in
N W7 using Adv_Fil e devices


Hey Calvin,

I can see how Adv_File devices would be usefully in smaller
environments, but it kinda doesn't seem very "Advanced" when you
consider the needs of larger installs.  I'm by no means a "large" site,
but I backup 200 clients and 8TB weekly, I would need a ~20TB FS just to
keep 2 weeks of backups "online".  Not to mention that I don't think one
array can handle the IO I would be generating for that many clients.  I
love the idea of being able the read and write at the same time, but why
does that have to be combined with the never getting full "feature".  I
would much rather have 20x1TB FSs then if one fills up it just moves the
save streams currently running to that Adv_File device over to another
device.

Maybe I'm just not understanding it correctly, but it seems like a
strange limitation.

Legato personnel feel free to jump in a correct my understanding of
Adv_File devices.

Thanks,

Joel

-----Original Message-----
From: Thomas, Calvin [mailto:calvin.thomas AT NACALOGISTICS DOT COM]
Sent: Friday, May 23, 2003 5:56 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] Initial Report on Functionality improvements in
N W7 using Adv_Fil e devices

The idea is you only use one Adv_file device, and it is never marked
full
even if it is full.  When NW7 detects the Adv_File device is full it
waits
for 10 minutes, during which it looks for expired save sets on the
Adv_File
device that have expired and it deletes them.  If it frees up enough
room at
the end of 10 minutes, the backup continues.  This makes a lot of sense
if
you change your expiration date so your save sets expire before filling
up
the Adv_File device. IE, in my case, I have 300 GB, and my daily backup
is
about 50GB (compressed).  I am going to set the Expiration time to 3
days,
so I have 150GB of un-expired save sets and 150GB of save sets on the
Adv_File device at any one time.  This allows me to keep the File device
full at all times for maximum recoverablility.  What about long term
storage?  I submit that Tape devices should ALL be clone devices now.
Every
save set should be automatically cloned to tape at the end of the
backup.
Here is my thoughts on the matter.

The Pro Side.
1. If the file device dies, or a save set is accidently deleted on the
hard
drive, you have everything on tape.  This is superior to staging alone
because with staging the save sets do not exist anywhere on tape unless
you
clone also (see 5 below).
2. The tape cloneing and backups happen simultaneously for more
flexibility
in scheduling.
3. No seperate tape Clone, and Backup pools. (less administration).
4. No worrys about waiting for staging to finish in time to recover the
space for your next backup to happen.
5. No worrys about the clone failing because the Adv_File device is busy
staging and can't clone and stage at the same time.

The Con Side.
1. I am not sure how long networker will store the indexes after the
save
sets are marked recycleable.  I read that it will keep the save set
information until the save sets are overwritten, and if there is a clone
on
tape, the indexes should be still be available, but I will feel better
about
it when I see it for myself.
2. If 1 above it wrong, the save sets would have to be scanned in.  Not
a
pretty thought.

Additional Thought:
Perhaps automatic expiration and staging could work together.  Perhaps
if
certain save sets need a LONG retention period, they can be staged while
the
rest are cloned and expired. It certainly gives me something to think
about
as I continue testing NW7.

Calvin E. Thomas
UNIX System Administrator
NACA Logistics




-----Original Message-----
From: Joel Fisher [mailto:jfisher AT WFUBMC DOT EDU]
Sent: Thursday, 22 May, 2003 01:47 PM
To: NETWORKER AT LISTMAIL.TEMPLE DOT EDU
Subject: Re: [Networker] Initial Report on Functionality improvements in
NW7 using Adv_Fil e devices


========(snip)================================================
One question I have along these lines is about the adv_file devices.  It
seems a bit odd to me that a save stream will not move to another volume
when the one it is currently writing to is full.  In a large environment
I would think that could cause some problems, you'd have to have
very(TB+) large volumes.  Anyone know what the idea behind these is?
I'd rather have 20x1TB adv_file devices over 2x10TB devices, but the
smaller the device the more likely it'll run out of room and just hang
there until something expires.  I don't have 7 yet so maybe this isn't
really a problem, but it seems like it could be.


Joel

--
Note: To sign off this list, send a "signoff networker" command via
email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=