- From: Julio Kriger <juliokriger AT gmail.com>
- To: Michael Tuexen <Michael.Tuexen AT micmac.franken.de>
- Cc: discussion AT sctp.de
- Subject: Re: [SCTP-Discussion] gap_report
- Date: Mon Jun 13 19:52:00 2005
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=nKzl3LXTl6svSKJrs7Xbese2OxRhQLDgW4EM/bu91fHgpRD2FHSN3vMl8wYWsR2sJnH5kOkLoor+eWpRmDmGjih9Mg+KjH6zklHhR7uY/ccHDs2KnVOESZGcZkvUrGakOH42YPJyY/OKHN8ZNYZMMbSvps2zVkOLE43t+2/wfgQ=
- List-id: SCTP Discussions <discussion.sctp.de>
Hi Michael!
Well, that is in what I'm working on.
The enhancements works with duplicate and missing reports.
I know that after 4 missing reports, a tsn get fast retransmitted.
Think about having statistic data on missing reports and duplicate
reports. So if you learn that the network don't loose packages, but
somehow, delay them, so that you receive an average of 5 or 6 missing
reports and the corresponding duplicate reports. Then you can change
that "4" of fst retransmit to 5 or 6 to get better performance because
yo will not retransmit fast retransmit spurios tsn. This will enhance
the throughput of sctp working on wireless networks. I hope :)
So I think I will have to add my own missing report counter.
Regards,
Julio
On 6/13/05, Michael Tuexen
<Michael.Tuexen AT micmac.franken.de>
wrote:
>
Hi Julio,
>
>
see my comments in-line.
>
>
Best regards
>
Michael
>
>
On Jun 13, 2005, at 5:47 PM, Julio Kriger wrote:
>
>
> Hi!
>
> I'm working on some enhancement of SCTP, specifically working on
>
> missing and duplicate reports.
>
> Browsing the code, I saw that de missing reports (in the code it is
>
> gap_report) after reaching to 4, it is set back to 0. Does anyone know
>
> why? TIA.
>
Why not? Once they have been scheduled for fast retransmit?
>
>
>
> Another question, have anyone try netem and this sctp library
>
> toghether?
>
No. But with dummynet on FreeBSD. That combination works.
>
>
>
> Regards,
>
> Julio
>
>
>
> --
>
> ----------------------------
>
> Julio Kriger
>
> mailto:juliokriger AT gmail.com
>
> _______________________________________________
>
> discussion mailing list
>
> discussion AT sctp.de
>
> http://www.sctp.de/mailman/listinfo/discussion
>
>
>
>
>
>
--
----------------------------
Julio Kriger
mailto:juliokriger AT gmail.com
Archive powered by MHonArc 2.6.19+.