DPDK usage discussions
 help / color / mirror / Atom feed
From: "Wiles, Keith" <keith.wiles@intel.com>
To: Alex Kiselev <kiselev99@gmail.com>
Cc: users <users@dpdk.org>
Subject: Re: [dpdk-users] a packet generator with traffic flow support
Date: Mon, 20 Feb 2017 21:23:04 +0000	[thread overview]
Message-ID: <136C2330-D66C-4F17-8F29-E6F3D6907B5A@intel.com> (raw)
In-Reply-To: <CAMKNYbw3sz4RpNSfurJ+MQw7w87qdoDWLGM8EbgVngw3PbRNiA@mail.gmail.com>


> On Feb 20, 2017, at 12:53 PM, Alex Kiselev <kiselev99@gmail.com> wrote:
> 
> Hi!
> 
> I am working on some stress tests for my packet forwarding engine
> and trying to figure out which packet generator tool can help me
> to accomplish my tasks with minimum efforts from me. What I need is something
> like PktGen but with a kind of traffic flow/session support.
> The flow/session in my case is a traffic flow with the following properties:
> - QinQ vids, src ip, dst ip etc
> - time to live
> 
> In addition, it would ideal if a tool could support defining pps rate of each
> flow or just a total pps of all flows.
> 
> So, a packet generator should be able to start in parallel some number
> of flows, then it should replace completed sessions whose ttl is
> expired with new
> ones until the goal (the given number of sessions or the given number
> of packets)
> is complete.
> 
> Maybe Warp17 is more close to the kind of tool that I need in terms of
> flow support, but I don't need tcp/udp flows and I need a way to
> define L2 packet headers.
> 
> Is it possible to code the flow logic that I need using PktGen
> scripting language?

The tool pktgen-dpdk does some of the items you want, but not all. Have you looked a the different modes pktgen-dpdk can handle like pcap or range support?

Also T-Rex traffic generator may have some of the features you are looking as it is a bit more programable using Python. Pktgen Lua scripting is really used to control Pktgen and not generate packets on the fly. Pktgen has a number of modes single, range, random, PCAP, sequence, which could be used with an external switch to merge the different TX options to a single target. I assume using an external switch maybe not work very well here.

> 
> I would appreciate any ideas. Thanks.
> 
> -- 
> --
> Kiselev Alexander

Regards,
Keith

  reply	other threads:[~2017-02-20 21:23 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-20 18:53 Alex Kiselev
2017-02-20 21:23 ` Wiles, Keith [this message]
2017-02-20 21:38   ` Alex Kiselev

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=136C2330-D66C-4F17-8F29-E6F3D6907B5A@intel.com \
    --to=keith.wiles@intel.com \
    --cc=kiselev99@gmail.com \
    --cc=users@dpdk.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).