perfSONAR User Q&A and Other Discussion

["Garnizov, Ivan (RRZE)" <>]

Hi Devine,

 

Which perfSONAR toolkit are you using? It does not become clear what exactly is the problem that you experience or you are in doubt with the toolkit you are using? Generally we are already using for several months iperf3 in production.

I can only tell you about iperf and I have no idea about nfdump tool. With this tool and your test you are measuring the TCP packets (real data) transferred.

 

Please use this link https://www.google.de/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#q=achievable%20throughput and check the third result (there are also multiple publications on the topic):

 

“For example, the maximum frame size in Ethernet is 1526 bytes: up to 1500 bytes for the payload, eight bytes for the preamble, 14 bytes for the header, and four bytes for the trailer. An additional minimum interframe gap corresponding to 12 bytes is inserted after each frame. This corresponds to a maximum channel utilization of 1526 / (1526 + 12) × 100% = 99.22%, or a maximum channel use of 99.22 Mbit/s inclusive of Ethernet datalink layer protocol overhead in a 100 Mbit/s Ethernet connection. The maximum throughput or channel efficiency is then 1500 / (1526 + 12) = 97.5 Mbit/s, exclusive of the Ethernet protocol overhead.”

 

Best regards,

Ivan Garnizov

 

 

 

From: [mailto:] On Behalf Of Tshabalala, Devine (Mr) (s210051701)
Sent: Dienstag, 6. Oktober 2015 14:14
To:
Cc: Szymon Trocha; Shukri Wiener; Roderick Mooi
Subject: [perfsonar-user] Iperf : Bandwidth clarification request

 

Good day

 

Hoping all is well, l am doing some bandwidth testing in a lab environment using iperf to determine the maximum achievable throughput on a 100Mbps link. I tested using the 64Kb windows size and got an average throughput of 82.7Mbps and l also used a window size of 96Kb which produced an average of 94.8Mbps. My question is , Why am l not achieving the full 100Mbps? I then ran again the iperf -w 64k and iperf - w 128Kb for 5 minutes to be able to collect the flow information using nfdump tool. On the console window l received almost similar results but the nfdump tool showed me different results for same window size . This gave rise to the second question of Why does the Nfdump indicates an improved iperf throughput compared to that shown on console window?. Lastly are there any formulas that are used to determine the achieve throughput that the iperf tool can produce?

 

The below are snapshots of my setup and the iperf testing done. Thank you

 

 

Kind regards

DevineTshabalala

Masters Student

Department School of ICT : l.T. Communication Networks

(North Campus)

Tel: 078 491 6130

 

A successful man is one who can lay a firm foundation with the bricks others have thrown at him and

Action is the foundational key to all success.