Connections

Networked services are an integral part of how a connected world operates.  We use them every day, often without realizing.  But our use of services can create a lot of waste information, a lot of which is private. This private information often accumulates in servers.  While some services find uses for that information, those uses can be disrespectful of individual privacy.  Oblivious HTTP is a new lightweight protocol that can be used to minimize the amount of private information that servers are exposed to.

Routing of IP packets is well-known to be based on distributed algorithms that select the next hop
towards the destination address along the best path to the destination. The measure of “best” is
based on the metrics associated with the link. Typically the algorithms that calculate the best path
are known as “shortest path first.”
Traffic Engineering (TE) techniques were developed to provide more sophisticated control of the
paths that collections of packet flows took through the network to guarantee bandwidth and steer
traffic away from the shortest paths.
Upper-layer applications (such as multiplayer immersive gaming, remote surgery, holographic
conferencing, or even haptic augmented reality multiplayer 3D worlds) are now placing increasingly sophisticated demands on the network for better quality, more predictability, and increased reliability. At the same time, lower-layer network technologies are advancing rapidly, providing increased bandwidth to the home and to mobile hand-held devices creating an environment that enables the potential of advanced applications running concurrently by vast numbers of end-users. These demands coincide with a growing trend to extend end-to-end communications to include machines and services through 5G, IoT and vehicular connectivity, and space-terrestrial communication.
The convergence of these innovations places requirements on the network to deliver particular
service level behaviours on a per-flow basis. Unfortunately, TE techniques for bundling extensive
collections of flows are neither scalable nor sufficiently granular to meet these needs.
Semantic Routing is the process of achieving enhanced routing decisions based on semantics added to IP headers to provide differentiated paths for different packet flows distinct from simple shortest path first routing. The additional information or "semantics" may be placed in existing header fields (such as the IPv6 Traffic Class field or the destination address) or may be achieved by adding fields to the header.
This talk will examine the application of Semantic Routing and how it allows packets from different flows (even between the same applications on the same devices) to be marked for different treatment in the network.
We will present an overview of some of the many ways that Semantic Routing has been proposed, from some well-established techniques to newer proposals such as SRv6 Network Programming and Application Aware Networking (APN). We will also examine how Semantic Routing may be integrated with SDN and programmable forwarding planes.
After a short discussion of the challenges and risks to the routing system associated with Semantic Routing, we will conclude with a description of the research challenges of Semantic Routing and an overview of where the research is being discussed.

Computation in the Network (COIN) is a wide ranging term as well as the name of an RG effort in the IRTF. This presentation aims at providing some insights into what COIN may be, where it may be headed and what work it may entail. Those insights are based on individual participation in the COIN RG with some example activities to underlie the key points. The presentation will also utilize the ongoing use case draft efforts to relate to some aspects of COIN.