Cisco Support Community
Showing results for 
Search instead for 
Did you mean: 

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

IPERF throughput comparison

I am currently using catalyst 4506 L3 switch and 2950 L2 switches for pc's which are EOL. I am replacing L3 with 3750X, two switches in a stack. L2 is 2960S all with gbps ports. I have to have a baseline (performance comparison with the current setup and new setup) and I am using the following methods

1. Iperf throughput comparison among local vlans

2. Iperf throughput comparison with remote data centers servers.

3. Ping/ftp upload/download comparison with remote DC's. FTP file is around 3GB.

Today I have tested the above with the new setup and compared it with the old setup. For the tasks 1 & 3 I found better performance with the new setup. However the task 2 did not show any performance increase infact the throughput was much lower comparing to the old setup. What is causing this behavior?

Appreciate your inputs.

New Member

Re: IPERF throughput comparison

I think you need to provide some more information. Please can you provide a logical (L3) diagram showing how the remote DCs are connected. I assume they are in different subsets, but this is not immediately clear? Also have you confirmed that all switchports are running at the same speed and duplex in the new scenario. show int status would help to show this.

Sent from Cisco Technical Support iPad App

New Member

Re: IPERF throughput comparison

Just to check if I understand your issue correctly...

I would assume that tests 2 and 3 would produce comparable results, because you are roughly testing the same connectivity, only with different tests-tools. But one test (iPerf) tells you performance degraded, and FTP tells you performance improved...

1. Using iPerf, what kind of test did you do (what was the commandline you entered).

2. You were using the exact same devices for these tests before and after the upgrade ?

3. And you were testing comparable times of the day (well I mean... the load of the network and servers was not an influencing difference between teh 'before and after' results)

New Member

Re: IPERF throughput comparison

Sorry for the trouble. I found the issue, for the "before test", I had used the laptop with 2GB ram and with new setup by mistake I used the laptop with 1GB ram and thats why the difference.