Amanda-Users

Re: Would like to get list's impression on (amanda 4TB backup)

2006-05-16 11:26:37
Subject: Re: Would like to get list's impression on (amanda 4TB backup)
From: "Lawrence McMahon" <lmcmahon AT lmcmahon.acsu.buffalo DOT edu>
To: Gordon.Mills AT usa DOT net
Date: Tue, 16 May 2006 11:22:09 -0400
  Hope it's ok if I jump in here.
I work at a University on our backups. I agree with everything Gordon said about TSM. We have both Amanda and TSM here. Our TSM setup is pretty much what Gordon described. We use Amanda to only backup what would be required to restore the system on a server. Like /, /var, and /opt (and any special files the user would require). We back up everything else with TSM. So if there were a disaster, the Admin for the server would use Amanda to restore the server, and then TSM to restore everything else. If an Admin just wants to restore a file, or some files, they use the TSM GUI like Gordon mentioned and do their own restores. Larry McMahon

Gordon J. Mills III writes:
Josef, here are some of my comments about TSM that I sent to a lister off
list. I think in a later email I will try to list the things that I liked
about TSM that I would like to see in Amanda. Some may not be possible.
At my previous job we used TSM and it is a very nice backup solution. We had
a tape library that held about 250 tapes and up to 8 drives (We had 5 drives
in it). It seems to excel at backing up large amounts of data. One unique
thing it does is that it uses and "incremental forever" backup strategy. It
backs up a machine fully only once (it can probably be setup other ways, but
that's the way we used it)<--(I could be wrong about that). There is a
backup server (similar to the way Amanda is setup) and it has a database
that keeps track of everything that has been backed up (all files). It knows
what files are on what tapes (Also similar to Amanda).
So instead of a traditional setup where you would backup fulls on Sunday
night and differentials or incrementals the rest of the week (or something
similar) and keep 4 sets of fulls for 4 weeks of history it has some
different settings. You set some parameters for each machine that is backed
up (along with the system default values).
You set a maximum time that it will keep old copies of files. (ex. 1 year)
You set the maximum number of copies it will keep of any single file. (ex.
30 copies).
You set the minimum number of copies it will keep of any single file. (ex. 5
copies) (Actually I am not certain about this setting)
There are many other settings too, for example how many copies of a deleted
file does it keep.
With the above parameters it will keep up to 1 year of file versions, but
not more than 30 copies and not less than 5. If the file changes every day
you will have 30 copies of it but only 1 month of history. If it changes
once a month you will have 12 copies/1 years worth.
It also keeps track of each tape's "fragmentation". Say you set it to 70%.
As files expire they will still be on a tape until overwritten. If a tape
has too many expired files on it, it will exceed the fragmentation
threshold. Once that happens the system will load the tape(s) and extract
the "good" data from it and write it to a new tape. Then the old tape(s)
will be put back into rotation for subsequent backups.
It can also be setup to account for offsite backups. Our tape library had a
hopper in it that was about 10 tapes large. It would figure out what files
were new and not yet at the offsite facility and write copies of all new
files to new tapes. It would put those in the hopper for the morning person
to collect. It would also print out a paper that listed the tape numbers
going offsite and which ones to bring back from offsite. Those would get put
back into the library.
With this method we always had a complete copy of everything ever backed up
in the offsite location. It also had a copy of everything it ever backed up
in the library. Obviously this can only work if you have a library large
enough to hold all of the data it will accumulate over a year or so.
There are 2 interfaces for the system, a web/java interface and a regular
application interface. The nice thing about the web interface is that you
can log on from any machine and tell it to restore a file to any client
using only a web browser. When restoring it can list all available versions
of a file or you can just set a date and tell it to restore to that date. Or
just take the default of the newest copies of all.
It can be setup to backup over whatever network you have. Ours had a 2Gb
optical SAN network for the IBM database servers and the diskdrives were
connected directly to that SAN. For the Windows clients (the ones I was
managing) they just backed up over the regular network (100Mb). It would
(similar to Amanda) backup over the network to a holding disk on the server
and then later push it out to tape. Very nice.
I believe the total cost of the system we had was around $500,000.00. Pretty
expensive, but a heck of a system! That included the tape library, the
backup server (I think it was an IBM AIX R6000 server), And the software. I
think I may have shocked the initial person requesting info on this with
this value. Please keep in mind that this was a lot of hardware and that was
the majority of the cost. The cost of the software is in line with other
commercial products IMHO.
After dealing with several different backup solutions, TSM is by far the
most impressive I've seen. When I started looking into Amanda at my new job
it seemed to have some of the features of TSM or at least was going in the
direction of the approach that TSM takes. For example it determines when to
make fulls, etc and all machines are not backed up as fulls on the same day.
Certainly an impressive system for a GPL project. And since it is used
mainly by people without mega-hardware, it probably wouldn't benefit the
users that much to take the approach of TSM. I know there must be
settings/etc for a much smaller implementation of TSM, I am just describing
what we had and how it worked. If I have not been clear on something I will try to explain better.
Also, please do not think I am badmouthing Amanda, I am not. I think it is a
fine system. Also, I am not a representative of IBM or TSM, I am/was just a
happy user at my old job.
Since Josef asked that I post to the list in hopes of generating a
discussion on some of the features of TSM and if it might be something to
add to the amanda wishlist, I would love to discuss that.
I hope that wasn't too much rambling. I just wrote down things as they came
to mind :-).
Regards,
Gordon Mills
-----Original Message-----
From: owner-amanda-users AT amanda DOT org [mailto:owner-amanda-users AT amanda DOT org] On Behalf Of Josef Wolf
Sent: Sunday, May 14, 2006 4:20 PM
To: amanda-users AT amanda DOT org
Subject: Re: Would like to get list's impression on (amanda 4TB backup) On Thu, May 11, 2006 at 12:34:00PM -0500, Gordon J. Mills III wrote: > Hello Ronald and Ross. Just a note on other backup solutions...If you > do decide to to commercial I would recommend you take a look at IBM's > Tivoli (aka TSM) backup system if you are looking at large libraries > and lots of data. It is a very impressive system with a completely > different stategy than other backup solutions I've seen. I'm very interested about the startegy. Would you please spend a couple of words about it to us? > That said, Amanda comes the closest so far to bridging the gap > (commercial or GPL). Amanda is a great backup solution, but if I > wanted to go commercial I'd definitely look at TSM. We had it at my old job and it was fantastic. What makes it superior to amanda?
> I
> won't go into a detailed description unless others would like to read > it. If you want to know more about the comparisons I will answer privately. I definitely would like to know more. And I think such a discussion would be perfectly on-topic on this list. There's no reason to go privately because most people on this list are interested (IMHO). > PS: I am not trying to start a flame war here nor am I an affiliate of IBM. > Just trying to give some info..... Hey, that's not fair! Please avoid statements like "they are superior" without an explanation _why_ they are superior. instead, please provide us with information about what is missing to come close to this superior system.




Lawrence McMahon
Senior Programmer Analyst
341 Computing Center, North Campus
645-3579

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