Testing Network Speeds

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

welchwerks

New Member
Feb 20, 2011
23
0
0
Finished a new server build with 08 with hyper v and 2011 guest (sorry doing this on my phone) I want to test transfer speeds from 2011 to workstations in the office. what do others use. I can explain my equipment later but for now it is all gigabit wired cat 6
 

apnar

Member
Mar 5, 2011
115
23
18
For simply testing network speed between boxes I'll usually use iperf (command line) or jperf (GUI). Which you can find here.
 

Patrick

Administrator
Staff member
Dec 21, 2010
12,511
5,792
113
Really depends what you are trying to model. Intel NASPT is very easy to use. One other thing that is "decent" is DUMeter. I run this all the time on my Windows 7 desktop.
 

SantaSCSI

New Member
May 22, 2011
20
3
3
Mine might not be as accurate as others, but I always test with a 8GB file by just copying it over CIFS/NFS/FTP (whatever you have enabled or want to use). It is a quick and dirty real-life test to check the every day speeds.

If you like more details, NASPT and indeed iperf are the way to go.
 

welchwerks

New Member
Feb 20, 2011
23
0
0
O.K. I used DU-Meter stopwatch and tranfered a 8.8 Mb mp3 from music shares folder on my desktop to desktop my Maximum transfer rate was 67.5 Mbps Average says 17.4 Mbps @ 4.2 seconds
A second transfer of 4.2 GB WHS 2011 iso file Maximum transfer rate 353.2 Mbps Average tranfer rate 49.7 Mbps @ 12 minutes 3 seconds

whats your thoughts?
 
Last edited:

murtoz

New Member
Aug 10, 2011
3
0
0
Wiltshire
Sounds like your limitation is disk speeds not network speed. I have a Gbit network here and get between 60 and 90 MByte/s sustained depending on which system is doing the writing. This is as determined by real life transfers of either movies (between 1 and 4GB on average), or music files (10MB or so on average). Oh and this is all on actual physical systems, not VM's. As a rule btw, the bigger the file, the higher the transfer speeds (sequential reads/writes being the cause of this; much more overhead for lots of little files)
All your figures are in Mbits/s not MBytes/s so a lot slower. Not sure how the virtual nature of your setup influences this. Also, I have jumbo frames enabled on all my gbit nics, set to the highest common value. Prior to enabling this I was not getting more than about 50 MByte/s.