Thanks for the info Toke. By running close to Max, I mean I want to load test it for a sustained period of time and measure the performance. I don't really have a clear idea specifically what I want to do, I just want to get some measurable metrics to be confident the line is working optimally, and have a baseline to compare to.
On E1 lines I used to do a loopback with a BERT test, to ensure no errors when running at 100%, 10G Ethernet doesn't really have an equivalent, so I am just making some stuff up. Previously I have just run a few iperf sessions, but I am hoping Flent can help us standardise a commissioning test for new circuits, and enable us to test and compare to the baseline under fault conditions. Regards Rich -----Original Message----- From: Toke Høiland-Jørgensen <t...@toke.dk> Sent: 28 March 2020 20:05 To: Hall, Richard <richard.h...@gsacapital.com>; flent-users@flent.org Subject: Re: [Flent-users] New 10G Pre-Provision testing ** This is an external e-mail. Please treat attachments and links as potentially dangerous. ** "Hall, Richard" <richard.h...@gsacapital.com> writes: > Hi everyone. I have recently discovered Flent. I am looking for some > guidance on how to use it for testing some new 10G transcontinental > circuits before I put them in service. I would like to run it close > to maximum capacity for a period of 24-48 hours to have confidence in > it's capability and reliability. I do not have any QOS or traffic > shaping to test, it is purely from a bandwidth, latency, jitter, MTU > and reliability perspective. Any other line provisioning advice would > be good to hear. Oh boy. I am not sure running a single Flent test for that long is going to work very well. It will probably run, but Flent is not really doing anything to limit its memory usage, or do any kind of stream processing, so there's a good chance it'll just gobble up all available RAM, then die. However, a way around this may be to just run a long repetition of shorter tests. 10 minutes each, say, then just repeating for as long as you like. That will get you test data in nice time-delimited chunks as well, which will probably make it easier to sift through the data. I expect you could just use the batch facility to do this; I've certainly run week-long test batches before. The details of how best to go about it will depend a little bit on what characteristics you wish to measure. You could just run the rrul_be test, I suppose. That'll load up the link with TCP traffic and measure latency as well (and all the TCP statistics you can pull out of the kernel). Is that sufficient, do you think, or did you have something more specific in mind? Or to put it another way, what exactly do you mean by "run it close to maximum capacity for a period if 24-48 hours"? :) -Toke For details of how GSA uses your personal information, please see our Privacy Notice here: https://www.gsacapital.com/privacy-notice This email and any files transmitted with it contain confidential and proprietary information and is solely for the use of the intended recipient. If you are not the intended recipient please return the email to the sender and delete it from your computer and you must not use, disclose, distribute, copy, print or rely on this email or its contents. This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. Any comments or statements made herein do not necessarily reflect those of GSA Capital. GSA Capital Partners LLP is authorised and regulated by the Financial Conduct Authority and is registered in England and Wales at Stratton House, 5 Stratton Street, London W1J 8LA, number OC309261. GSA Capital Services Limited is registered in England and Wales at the same address, number 5320529.
_______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
