I use Storage Unit Groups and used to use "Any Available" but IMO that's
a sad representation of load balancing akin to "round robin" DNS load
balancing.  What I would like to see is NBU limiting the number of jobs
that a device can handle per max speed as opposed to per number of
streams as it is now.  If I had two LTO3 100MB/sec drives w/ 4
concurrent mpx and I kicked of 8 jobs in the order of GigE / 100F / GigE
/ 100F the 4 GigE jobs would be running on the same drive at 25MB/sec
and the 4 100/F jobs would be running at 40MB/sec total on a drive that
can accept another 60MB/sec.  I know this is a gross over
simplification, but I don't think its a terribly hard concept to
implement.  Replace the current Max Streams for Storage Unit (DSU/DSSUs
included!) and replace it with a Max MB/sec and you'll see real load
balancing.  I know there's technical hurdles (like jobs not running at
consistent speeds) but that's what Symantec pays software engineers for
;)
 
-Jonathan
 


________________________________

From: veritas-bu-bounces at mailman.eng.auburn.edu
[mailto:veritas-bu-bounces at mailman.eng.auburn.edu] On Behalf Of Curtis
Preston
Sent: Monday, May 14, 2007 6:43 PM
To: Steven L. Sesar
Cc: VERITAS-BU at mailman.eng.auburn.edu
Subject: Re: [Veritas-bu] Newbie to Netbackup using VTL



If you use a storage unit group (available for a few versions now) or
"any available," any available media server can backup any other media
server's client, or at least any clients that are configured that way.

 

Still don't have an answer to your other question. ;)

 

---

W. Curtis Preston

Author of O'Reilly's Backup & Recovery and Using SANs and NAS

VP Data Protection

GlassHouse Technologies

 

________________________________

From: Steven L. Sesar [mailto:ssesar at mitre.org] 
Sent: Monday, May 14, 2007 12:18 PM
To: Curtis Preston
Cc: Meidal, Knut; VERITAS-BU at mailman.eng.auburn.edu
Subject: Re: [Veritas-bu] Newbie to Netbackup using VTL

 

Curtis Preston wrote: 

Steven L. Sesar said:
 
  

        The way that CommVault does it with Gridstor is that each media
server 
        shares an index cache (the equivalent of NBU's
catalog/databases) via 
        NFS or CIFS. I call it a "poor man's clustered filesystem". It
affords 
        one the luxury of pooling media servers, such that if one goes
down,
            

any 
  

        other can take over in it's place. Since the index cache is
shared, the
            

 
  

        job just picks up where it left off. It also allow one to load
balance 
        among media servers. 
            

 
NBU's index is already centralized at the master (and therefore shared
amongst media server), so media servers can take over for other media
servers any time.
  


Really? They will automatically failover/load balance? Something new in
6.x? How are the media servers logically grouped? Is this now a function
of STU groups? (pssst! I'm the guy who asked you the question at SNW
about email archiving's potential impact on de-duplication - still
haven't gotten a good answer on that one)




 
  

        The same goes for disk storage targets. If disk 
        storage unit "A" becomes unavailable for any reason, disk STU
"B" can
            

be 
  

        configured to take over in it's place. Massive job and component
            

resiliency!
 
This is where CommVault shines.  They can share a disk device on the SAN
where NBU can't (yet), and backups can span disk devices where NBU can't
(yet).  NBU 6.5 (due next month) is supposed to have both of these.
  






-- 
===================================
 
   Steven L. Sesar
   Lead Operating Systems Programmer/Analyst
   UNIX Application Services R101
   The MITRE Corporation
   202 Burlington Road - MS K101
   Bedford, MA 01730
   tel: (781) 271-7702
   fax: (781) 271-2600
   mobile: (617) 519-8933
   email: ssesar at mitre.org
 
=================================== 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: 
http://mailman.eng.auburn.edu/pipermail/veritas-bu/attachments/20070515/8d6d807e/attachment.html