speaker: Marcelo RIcardo Leitner (Red Hat), Xin Long (Red Hat)
SCTP is a transport protocol, like TCP and UDP, originating from SIGTRAN
IETF Working Group in the early 2000's with the initial objective of
supporting the transport of PSTN signalling over IP networks. It featured
multi-homing and multi-stream from the beginning, and since then there
have been a number of improvements that help it serve other purposes too,
such as support for Partial Reliability and Stream Scheduling.
Linux SCTP arrived late and was stuck. It wasn't as up to date as the
released RFCs, while it was also far behind other systems such as BSD,
and also suffered from performance problems. In the past 2 years, we
were dedicated to ensuring that these features were addressed and
focused on making many improvements. Now all the features from released
RFCs have been fully supported in Linux, and some from draft RFCs are
already ongoing. Besides, we've seen an obvious improvement in performance
in various scenarios.
In this talk we will first do a quick review on SCTP basics, including:
Background: Why SCTP is used for PSTN Signalling Transport, why other
applications are using or will use SCTP.
Architecture: The general SCTP structures and procedures implemented in
VS TCP/UDP: An overview of functions and applicability of SCTP, TCP and
Then go through the improvements that were made in the past 2 years,
SCTP-related projects in Linux: Other than kernel part, there are also
lksctp-tools, sctp-tests, tahi-sctp, etc.
Features implemented lately: RFC ones like Stream Scheduling, Message
Interleaving, Stream Reconfig, Partially Reliable Policy, and many
CMSGs, SndInfos, Socket APIs.
Improvements made recently: Big patchsets like SCTP Offload, Transport
Hashtable, SCTP Diag and Full SELinux support.
VS BSD: We will take a look at the difference between Linux and BSD now
regarding SCTP. You will be surprised to see that we've gone further
than other systems.
We will finish by reviewing a list of what is on our radar as well as next
Ongoing features: SCTP NAT and SCTP CMT, two big important features are
ongoing and already taking form, and more Performance Improvements in
kernel have also been started.
Code refactor: New Congestion Framework will be introduced, which will
be more flexible for SCTP to extend more Congestion Algorithms.
Hardware support: HW CRC Checksum and GSO will definitely make performance
better, for which a new segment logic for both .segment and HW that works
for SCTP chunks is needed.
RFC docs improvements: We believe that more extensions and revisions will
make SCTP more widespread.
For its powerfulness and complexity, SCTP is destined to face many challenges
and threats, but we believe that we have already and will continue to make it
better than that on other systems, but also than other transport protocols.
Please join us, Linux SCTP needs your help too!