Host-based backup of VMware vSphere VMs.
Post Reply
AlexWhit
Enthusiast
Posts: 40
Liked: never
Joined: Jul 21, 2011 10:10 am
Full Name: Alex Whittaker
Contact:

Network

Post by AlexWhit »

I wonder if you can help. i have just purchased a new server to run veeam. it is a dell T310 with quad core xeon 4gn ram 5tb free hard drive space.

then i am running the team backup i am getting bottleneck network (processing rate is 70 MB/s). the server is running windows 2008 R2 and is also hosting the vSphere server. the backup is going to be local then copied to usb3 HDD. I also have a backup proxy running. this is one of the servers on out VMware hosts with another 8 cores available.

is this to be expected on the speed or should i be getting quicker?

thanks
foggy
Veeam Software
Posts: 21069
Liked: 2115 times
Joined: Jul 11, 2011 10:22 am
Full Name: Alexander Fogelson
Contact:

Re: Network

Post by foggy »

Alex, quite fair speed for the hotadd processing mode actually. What are the full bottleneck analysis numbers you get? Is this processing rate measured for the full backup?
AlexWhit
Enthusiast
Posts: 40
Liked: never
Joined: Jul 21, 2011 10:10 am
Full Name: Alex Whittaker
Contact:

Re: Network

Post by AlexWhit »

yes this is for a full backup. ( the first on this server.) looking at the network card it is only really using 32% i may move it on tot he same switch as the servers to see if that helps.

where do i get the full bottleneck analysis numbers?
foggy
Veeam Software
Posts: 21069
Liked: 2115 times
Joined: Jul 11, 2011 10:22 am
Full Name: Alexander Fogelson
Contact:

Re: Network

Post by foggy »

It's on the line just above the one telling you the primary bottleneck (in the job real-time stats window).
AlexWhit
Enthusiast
Posts: 40
Liked: never
Joined: Jul 21, 2011 10:10 am
Full Name: Alex Whittaker
Contact:

Re: Network

Post by AlexWhit »

duration 3:07
processing rate 81 MB/s

source 30%
proxy 67%
network 80%
target 1%

is this what you were after?
Gostev
Chief Product Officer
Posts: 31459
Liked: 6648 times
Joined: Jan 01, 2006 1:01 am
Location: Baar, Switzerland
Contact:

Re: Network

Post by Gostev »

Yes, according to these number, the bottleneck here is clearly the network throughput. There is no point in looking at specific network card's network usage though - it shows percent used from maximum possible speed on totally empty network. However, remember that your network's bandwidth is shared between multiple servers, and there are lots of traffic from other servers too at any time. Thanks.
AlexWhit
Enthusiast
Posts: 40
Liked: never
Joined: Jul 21, 2011 10:10 am
Full Name: Alex Whittaker
Contact:

Re: Network

Post by AlexWhit »

Thank you for that. I just wanted to make sure i was getting the best possible speed for our backup
ThomasMc
Veteran
Posts: 293
Liked: 19 times
Joined: Apr 13, 2011 12:45 pm
Full Name: Thomas McConnell
Contact:

Re: Network

Post by ThomasMc »

Are your NICs broadcom?

I done a fair bit of testing NICs and found

broadcom = Bad
Intel = Great
Intel ET = Oh Yeah :D
AlexWhit
Enthusiast
Posts: 40
Liked: never
Joined: Jul 21, 2011 10:10 am
Full Name: Alex Whittaker
Contact:

Re: Network

Post by AlexWhit »

yep Broadcom.. i don't think it will be a major issue as the full backup will be done on a saturday night over the weekend. i think it will only take 10 hours for all virtual machines the worst bit will be copying the backup filed to the USB 3 hard drives that i need to figure out how to do correctly
Post Reply

Who is online

Users browsing this forum: No registered users and 80 guests