Session reports for SIGCOMM 2010
Computer communication review
This document collects together reports of the sessions from the 2010 ACM SIGCOMM Conference, the annual conference of the ACM Special Interest Group on Data Communication (SIGCOMM) on the applications, technologies, architectures, and protocols for computer communication. • Examples of the spanning tree did sub-optimal stuff: using just 2 Ethernets and 1 bridge. • Problems with bridges: "If you cannot keep up with the wire speeds, you are not allowed to be a bridge". • Views on why IP is
... imal ("why not replace bridges with IP routers?") -one should check DECnet and CLNP for this. The Boston hospital example: the company providing the switches used one huge bridged network. Bridging was never intended to do that: it was kind of a hack because people at the time were all confused about what Layer 3 was and they thought Ethernet was a competitor to DECnet. With bridges we did such a good job and it was so plug-and-play that you didn't have to think about them, so people are still taking large networks and doing bridges. On TRILL -Transparent Interconnection of Lots of Links • Concept of RBridges: devices that implement the TRILL protocol. • Gist given in the version 2 of Algorhyme A network where RBridges can Route packets to their target LAN. The paths they find, to our elation, Are least cost paths to destination! With packet hop counts we now see, The network need not be loop-free! About computing trees deterministically. With respect to supporting multiple interfaces (as questioned by an audience member), TRILL has problems with computing and choosing multiple trees that she wants students to solve. Motivation: Current applications leverage partially correct packets because some of these are recoverable into useful information; however some of them are corrupted to extent that they are not recoverable, which results in wastage. Key idea: The usefulness of such partially correct packets depend on the number of error being below a threshold such that useful information is recoverable, an Error Estimation can help in judging if packet should be forwarded or retransmitted. Internet Inter-Domain Traffic Craig Labovitz (Arbor Networks), Scott Iekel-Johnson (Arbor Networks), Danny McPherson (Arbor Networks), Jon Oberheide (University of Michigan), Farnam Jahanian (University of Michigan) Measuring the Internet is hard. There is limited "ground-truth" on inter-domain traffic. Objective: Seek to examine the changes in Internet inter-domain traffic demands and interconnection policies. Methodology: Leverage widely deployed Internet monitoring infrastructure and coax carriers into participation. Analysed more than 200 ExB of commercial Internet traffic over a two year period.