2,669MB/min on BE is fairly reasonable all in.
I tend to measure a point to point (2 systems) 1Gbit/s network as 'around' 60MBytes/s to be fair to it so that's just over 3,600MB/min flat out.
I think I'd just leave it be
Typically it'll go faster with multiple hosts, faster bus speeds, disks etc.
Cracking article if you want a read - http://www.tomshardware.com/reviews/...th,2321-3.html
Last edited by kmount; 25th August 2012 at 02:42 PM.
The network is 1gbit speed though. backup server has teamed network cards running at 2Gb/s ( Fixed the team issue earlier which seems to of speeded it up some) and file server is running at 1Gb/s and is a virtual machine - links to switches all 1Gb.
I'll have read of the article thanks
guess this first part on your link explains it. I thought was 1Gb was faster, silly me =]
How fast is a gigabit? If you hear the prefix "giga" and assume 1,000 megabytes, you might also figure that a gigabit network should deliver 1,000 megabytes per second. If this sounds like a reasonable assumption to you, you’re not alone. But unfortunately, you’re going to be fairly disappointed.
So what is a gigabit? It is 1,000 megabits, not 1,000 megabytes. There are eight bits in a single byte, so let’s do the math: 1,000,000,000 bits divided by 8 bits = 125,000,000 bytes. There are about a million bytes in a megabyte, therefore a gigabit network should be capable of delivering a theoretical maximum transfer of about 125 MB/s.
Last edited by ihaveaproblem; 25th August 2012 at 02:47 PM.
Depends how the team is created, if it's LACP it doesn't mean you'll get 2Gb/s between two points because of how it works.
You would find that if you were backing up multiple hosts from one back up server you will use more than the single 1Gb/s link because streams between A-B may go down nic1, but streams for A-C may be down nic2.
All depends what the teaming is really.
But I wouldn't be disappointed with the throughput you're getting especially from a VM if it's sharing resources with other stuff.
Though if it's a VM you may improve performance by talking to the hypervisor instead of the guest, depending on what hypervisor it is.
Ah no worries
It's Vmware esx 4.0
esx hosts connect to a hp MSA2312fc where all the file storage is kept.
On the backup exec selection all I've done is find the file server, and selected the d: drive to be backed up ( D: is all the file storage on the msa ) Is that the correct way to be backing up for my setup?
There's a few ways to do it.
You could continue doing it this way where you're talking to the guest and using the BE Remote Agent; or the other way is licensing BE for VMware and talking to the vcenter (or) hypervisor(s) and back up the whole VM from that way so you can restore a full VM.
AFAIK yeah, I think that's the functionality they refer to as GRT (Granular Recovery Technology).
ihaveaproblem (25th August 2012)
Do you happen to know how to link backup jobs together ?
I understand you can create policy's, but it seems like a policy has to use the same selection list.
What I want is the 1st job that's using selection list A to finish running, then job 2 using selection list B starts.
I've read this guide but don't really understand it
I'm sorry I don't, I don't use BE enough to have needed to do that.
I'm sure someone else will pop along and will know
Our backups started taking ages - nearly 2 days for one Win2008 system (and very slow MB/min)!
In the end we upgraded to the latest patch levels of BE for our particular version (which required the latest windows patches as well) and hey presto - back to six hours or so.
Check that you are on the latest patch levels of both BE and Windows.
WE had very slow backup jobs to disc. We recreated the backup to disc folder and everything fixed itself over the next few weeks.
There are currently 1 users browsing this thread. (0 members and 1 guests)