• GI/ITG KuVS Fachgespräch Future Internet
  May 28th, 2009 in Munich, Germany
  http://www.docomoeurolabs.de/FI/cfp.html
  Deadline: May 6, 2009

• 9th Würzburg Workshop on IP:
  Joint ITG and EuroNF Workshop
  “Visions of Future Generation Networks”
  (EuroView2009)
  July 27th – July 28th 2009 Würzburg, Germany
  http://www3.informatik.uni-wuerzburg.de/euroview/2009/
  Deadline: May 22, 2009

There had been some interessing talks especially about my favorite transport protocol SCTP and the IETF’s first session layer protocol “RSerPool” (IETF charta), which will allow you to build reliable server pools using SCTP. There are also implementations for both available from the University Duisburg-Essen.

Additionally my paper (slides, BibTeX) on a label-forwarding approach based approach was accepted

More information on the European Future Internet activities can be found here and here.

“The way we’ve always done it is not necessarily the same as the right way to do it”

The background is that the networking stacks we know today were initially created for MULTICS machines and had to had the well-known in-kernel design (networking card -> Interrupt service routine -> soft interrupt -> socket interface -> user space read()) because of the very slow paging mechanisms of MULTICS. This would not matter, if all computers would be singlecore machines. But on a multicore machines all the protocol work should be moved to the processor core that’s going to handle the data.

As a result a linear scalability on multicore systems can be achieved and everything gets a lot faster (about 6 times). Within the slides there are some measurements done by Van Jacobson on a patched linux machine supporting the benefits of this networking design.

But there is also some critisism from LWN here and here on this networking approach. Especially this detail might be of interest:

In traditional approach you also fetch all the headers on softirq,
but you do all the required work with them immediately and do not access
them when the rest of processing is done in process context. I do not
see how netchannels (without hardware classification) can improve
something here. At the first sight it makes locality worse. [...]
It is an amazing toy. But I see nothing, which could promote its status
to practical. Exokernels used to do this thing for ages, and all the
performance gains are compensated by overcomplicated classification
engine, which has to remain in kernel and essentially to do the same
work which routing/firewalling/socket hash tables do.

So more research might be neccessary to implement Van Jacobsons idea and make it a success.

“The challenges and and possibilities are out there, waiting for the next group of dreamers to change the world once again.”

I blogged about this election by the INTERNET WORLD Business newspaper some time ago, but now it’s your chance to support us. Just click on the following link and give us your vote

Update: The competition is also mentioned on the Biz-Award-Website and the Gründerzene-Blog.

The Internet, an undergrad project by UCSD graphic art student Jennifer Hsu.

(via: blog.caida.org)

As a real technology company SONES will not only design and implement new concepts for searching, but also new concepts for storing these information in application specific distributed filesystems (e.g. GoogleFS) using latest ideas related to Future Internet designs.

Today, while I’m visting the EuroView Workshop in Würzburg (Germany), we were noticed that the INTERNET WORLD Business newspaper has nominated SONES as one of the most important (german) start-ups. As a reader of this newspaper you can vote for SONES till 30. September 2008. The winners of the contest will receive an Onlinestar during the INTERNET WORLD Congress in Munich in October.