SCTP Discussions

Text archives Help


Re: [SCTP-Discussion] [an uncomfortable case when using sctptftp based on sctplib-1.0.8]


Chronological Thread 
  • From: Michael Tüxen <Michael.Tuexen AT micmac.franken.de>
  • To: 张文涛 <zhangwt1989 AT gmail.com>
  • Cc: sctp-discussion AT sctp.de
  • Subject: Re: [SCTP-Discussion] [an uncomfortable case when using sctptftp based on sctplib-1.0.8]
  • Date: Sun, 18 Oct 2009 14:14:25 +0200
  • List-archive: <http://lists.franken.de/pipermail/sctp-discussion>
  • List-id: SCTP Discussions <sctp-discussion.sctp.de>

Dear all,

unfortunately, we are not actively working on the sctplib anymore.
I would highly encourage you to use the kernel implementations of SCTP
for Linux or FreeBSD. They are both much better than the sctplib.

If you want to try the FreeBSD kernel implementation, please consider
using FreeBSD 8. For any problem regarding the FreeBSD implementation
I might be able to help.

Regarding your analysis:
* SCTP should not perform worth than TCP. The problems you see are related
to the sctplib implementation, not the protocol.
* After the retransmission of a chunk its size should be subtracted exactly once
against from the peer window, not twice. It it is subtracted twice (one time
at the first transmission, second time at the retransmission, and it is
not added when the chunk is scheduled for retransmission), it is a bug.
* Processing rtx->chunk_list might take a long time. The sctplib, in general,
has not been optimized for speed.

If you are providing bug fixes, we might integrate them in future releases.
But as said above, we are not actively maintaining it anymore, so it might
take some time...

Best regards
Michael

On Oct 18, 2009, at 12:23 PM, 张文涛 wrote:

Dear all,
We have met an uncomfortable case when using the example application “sctptftp” based on sctplib to transport files between two single-address peers. Throughput in this case is much worse than expected, and worse than TCP.
The specific circumstance is in the attachment.
Hope you can solve the problem presented in part 5(more in part 6,7), or clarify the problem and find the real cause.
Looking forward to your reply.
Wentao Zhang
E.E . Tsinghua University <email.doc>_______________________________________________
sctp-discussion mailing list
sctp-discussion AT sctp.de
http://www.sctp.de/mailman/listinfo/sctp-discussion





Archive powered by MHonArc 2.6.19+.

Top of page