BackupPC-users

Re: [BackupPC-users] Centralized storage with multiple hard drives

2014-03-20 15:00:09
Subject: Re: [BackupPC-users] Centralized storage with multiple hard drives
From: Timothy J Massey <tmassey AT obscorp DOT com>
To: "General list for user discussion, questions and support" <backuppc-users AT lists.sourceforge DOT net>
Date: Thu, 20 Mar 2014 14:59:02 -0400
Les Mikesell <lesmikesell AT gmail DOT com> wrote on 03/20/2014 01:59:47 PM:

> On Thu, Mar 20, 2014 at 12:21 PM, Timothy J Massey
> <tmassey AT obscorp DOT com> wrote:
> >
> > >
> > > You've used BackupPC before, Les, right? ;-)
> > > BackupPC prefers pool reads over writes when possible, and it typically
> > > accesses large amounts of data almost randomly. Caching metadatawill help,
> > > caching data likely won't.
> >
> >
> > I've given up explaining this to Mr. Mikesell.  I've posted info
> to the list several times showing the performance on BackupPC
> servers that had 512MB (Megabytes) of RAM.  Zero swapping, and
> something like 300MB used for caching.  I have then upgraded them to
> 4GB of RAM (8 times as much RAM!) and saw a whopping 15 or so
> minutes savings on a backup that took 10 to 12 hours.  (In fact, the
> only real reason I use 4GB minimum across the board now is that I
> ran into a problem where a fsck wouldn't complete without more RAM.)
>
> 4GB is still a tiny amount of RAM these days.  Try 20+.  You basically
> want to cache all of your pool directory/inode data plus the pc backup
> trees being traversed by rsync.  And without threshing it out with the
> rsync copy of the remote trees that are being processed.


After this, I *really* give up.

I *have* servers with 32GB of RAM.  I've gone from 4GB to 16GB and seen zero performance increase.

Do you *REALLY* think this is some sort of binary on/off enhancement?  Have 4GB?  No benefit.  5GB?  WOW Look at it go!!!   If you see zero improvement with 8 TIMES AS MUCH RAM (when you're not swapping to begin with!), you expect me to believe that double or triple that and all of a sudden a big difference?  Come on now...

> > And I haven't seen that RAID-5 has been that much of an issue.  As
> you mention, BackupPC is *READ* heavy so the RMW penalty doesn't
> hurt much.  My experience with both RAID-5 *AND* RAID-6 (even
> software RAID-6!) is just fine.
>
> Sure, they work.  But they force all of the disk heads to seek every
> time unless you have a large number of disks, and every short write
> (and most of them are with all of the directory/inode operations
> happening), is going to cost an extra disk revolution time for the
> read/modify/write step.  And that 'read heavy' assessment is somewhat
> optimistic if you have any large files that are modified in place.
> There you get the worst case of alternating small reads/writes in
> different places as the changes delivered by rsync are merged into a
> copy of the old file.


I'm not saying that there isn't a penalty for RAID-5 or -6.  I'm saying that the penalty is usually NOT RELEVANT FOR THIS APPLICATION!  Not in theory, but in ACTUAL PRACTICE.  You then go on to make my point:

> > Of course, I have a minimum of 4 drives in a RAID array (6 minimum
> for RAID-6), so I'm usually bottlenecked somewhere else anyway, such
> as a single 1Gb link.  It's not hard to mange writing 70MB/s of data!
>
> Seriously?  You have rsync pegging a 1Gb link for long periods of time
> anytime but the first run?  Do you have something creating a lot of
> huge new files?  My times are mostly constrained by the read time on
> the targets with a relatively small amount of data transfer.


You have stated that you're limited by the TARGET.  So given that, who cares if the disks are 60% busy instead of 30% busy while it waits on those targets?  And why WOULDN'T I use something like RAID-5 or -6 and give myself a LOT more usable space?

And my point was not that I *am* limited by my 1GbE link (I'm not usually with BackupPC but I *am* with other uses of my backup servers in general), but that at *BEST* I can move 70MB/s *because* I'm using 1GbE.  Seeing as my 6-drive RAID-6 array can copy files from one spot on the array to another (so not just simply streaming reads or writes) at 400MB/s routinely (and >1,000MB on my 12-drive systems), who *CARES* how much faster it would be if it were, say, RAID-10?  This server even has 4 x 1GbE, but you *never* get that with bonded Ethernet.  You're lucky to get 150MB/s.  So my drives are NEVER THE ISSUE!

So, yes, I'll take the extra space, which is *ALWAYS* my limiting factor!  :)

Tim Massey
 
Out of the Box Solutions, Inc.
Creative IT Solutions Made Simple!

http://www.OutOfTheBoxSolutions.com
tmassey AT obscorp DOT com
      22108 Harper Ave.
St. Clair Shores, MI 48080
Office: (800)750-4OBS (4627)
Cell: (586)945-8796

------------------------------------------------------------------------------
Learn Graph Databases - Download FREE O'Reilly Book
"Graph Databases" is the definitive new guide to graph databases and their
applications. Written by three acclaimed leaders in the field,
this first edition is now available. Download your free book today!
http://p.sf.net/sfu/13534_NeoTech
_______________________________________________
BackupPC-users mailing list
BackupPC-users AT lists.sourceforge DOT net
List:    https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:    http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/