Sid_F
asked on
Iperf output question
I am running iperf across a few different type of links. I have a server setup on the other end and I am getting varying values depending on the line. I'm finding the output very up and down. e.g iperf -c myserver -I 1 can produce speeds of anywhere from 30Mbps - 85mbps.
I am a little confused on the transfer figure. Why does the amount being transferred keep changing. I have looked at a number of videos but they don't seem to give specifics on the actual output. Sorry for the stupid question but I'm guessing that the bandwidth figure being presented is relative to the amount of data being transferred e.g I can't say my network has a 44Mbit/sec throughput to server X, I have to say when transferring 7.25MBytes per second my network has a throughput of 44Mbit/sec to the server..
I am a little confused on the transfer figure. Why does the amount being transferred keep changing. I have looked at a number of videos but they don't seem to give specifics on the actual output. Sorry for the stupid question but I'm guessing that the bandwidth figure being presented is relative to the amount of data being transferred e.g I can't say my network has a 44Mbit/sec throughput to server X, I have to say when transferring 7.25MBytes per second my network has a throughput of 44Mbit/sec to the server..
Instead of -I 1 do -I 60
That will give you the average bandwidth over 60 seconds
That will give you the average bandwidth over 60 seconds
Network congestion, QOS, server utilization... There are many things that can affect performance. If you are testing across a network that others are using, and testing a server that others are using, then you have to share resources with them. If this is a local network, and a dedicated test server, something else may be going on. What is the server's NIC, CPU, & Mem utilization when you are NOT running iperf?
Good catch, Aaron Tomosky. Why are you changing the sampling time? The default is 10 seconds, and should be good enough for a snapshot of a network without tying up the line for other people.
ASKER
Thanks all, It's a wired network, I'm using iperf to get familiar with my network benchmarks but with such a huge difference in speed its difficult to know what through put I have. If for a minute I suppose that a third party has indicated my network is the reason there application is slow, if I ran the throughput test I couldn't say hey I'm getting 72Mbits throughput as some of the below indicates as low as 26Mbits?.
As mentioned in my OP I am still confused on the transfer amounts, why does it transfer 8MBytes and then jump to transferring 72MBytes?
[ ID] Interval Transfer Bandwidth
[ 3] 0.0- 1.0 sec 8.62 MBytes 72.4 Mbits/sec
[ 3] 1.0- 2.0 sec 8.88 MBytes 74.4 Mbits/sec
[ 3] 2.0- 3.0 sec 8.12 MBytes 68.2 Mbits/sec
[ 3] 3.0- 4.0 sec 8.62 MBytes 72.4 Mbits/sec
[ 3] 4.0- 5.0 sec 8.50 MBytes 71.3 Mbits/sec
[ 3] 5.0- 6.0 sec 6.75 MBytes 56.6 Mbits/sec
[ 3] 6.0- 7.0 sec 8.50 MBytes 71.3 Mbits/sec
[ 3] 7.0- 8.0 sec 3.12 MBytes 26.2 Mbits/sec
[ 3] 8.0- 9.0 sec 5.25 MBytes 44.0 Mbits/sec
[ 3] 9.0-10.0 sec 5.75 MBytes 48.2 Mbits/sec
[ 3] 0.0-10.0 sec 72.2 MBytes 60.5 Mbits/sec
As mentioned in my OP I am still confused on the transfer amounts, why does it transfer 8MBytes and then jump to transferring 72MBytes?
[ ID] Interval Transfer Bandwidth
[ 3] 0.0- 1.0 sec 8.62 MBytes 72.4 Mbits/sec
[ 3] 1.0- 2.0 sec 8.88 MBytes 74.4 Mbits/sec
[ 3] 2.0- 3.0 sec 8.12 MBytes 68.2 Mbits/sec
[ 3] 3.0- 4.0 sec 8.62 MBytes 72.4 Mbits/sec
[ 3] 4.0- 5.0 sec 8.50 MBytes 71.3 Mbits/sec
[ 3] 5.0- 6.0 sec 6.75 MBytes 56.6 Mbits/sec
[ 3] 6.0- 7.0 sec 8.50 MBytes 71.3 Mbits/sec
[ 3] 7.0- 8.0 sec 3.12 MBytes 26.2 Mbits/sec
[ 3] 8.0- 9.0 sec 5.25 MBytes 44.0 Mbits/sec
[ 3] 9.0-10.0 sec 5.75 MBytes 48.2 Mbits/sec
[ 3] 0.0-10.0 sec 72.2 MBytes 60.5 Mbits/sec
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
ASKER
Thanks sorry for delay
If both end points are not loaded with any other services, and the network is stable, then the reading should be close.
If it's wireless in a busy environment, then all bets are off. In a wireless environment I can get 5Mbps, then 15Mbps only a few seconds later. Or by taking one step around a corner, or by standing instead of sitting...many variables.
What kind of environment are you testing in?